I'm happy to throw money into that. I produced the dCoreDuino with an inline i2c bus with exactly that in mind. I also think it's good for Arduino, but lets be transparent about it, money will be spent, time will be contributed, money may or may not be made, and at the end of the day, the main beneficiary will be Arduino and the opportunity sales spurned on by a range of i2c devices built with it in mind.
So I think, at the very least, they might provide some guidance about how these collaborative projects might best be organized, and how they might support a range of these devices.
All true, and I'm sure they're listening so let's see what they have to add.
One thing I'd hate to see happen is us to degenerate into who paid for it, who gets to profit from it.
And a question to all you Arduino/compatible product manufacturers and resellers - have you ever considered putting all your products into one site, rather than operating individual sites?
Have you ever considered that together it might be possible to build a single e-commerce entity that could rival the likes of Sparkfun?
That's true, but who's going to run the site, and would they have to pay for the stock of so many different products or would all the individuals give a supply of their designs as a consignment?
Interesting... a collaborative e-business. Implementing regional distribution centers make more sense, the difficult part is finding or writing the software to control all the operations. The e-commerce software is still very immature, and there is a huge opportunity for someone who wants to do it right... yahoo couldn't do it right, neither amazon.com...
The shipping would be deadly if you order 4 products from the site and they come from 4 different suppliers. It would have to be all items shipped from the same place.
Interesting... a collaborative e-business. Implementing regional distribution centers make more sense, the difficult part is finding or writing the software to control all the operations. The e-commerce software is still very immature, and there is a huge opportunity for someone who wants to do it right... yahoo couldn't do it right, neither amazon.com...
The shipping would be deadly if you order 4 products from the site and they come from 4 different suppliers. It would have to be all items shipped from the same place.
I'm fairly certain NKC was pushing for the opposite idea entirely. Currently, there are small-time sellers spread out everywhere, what if they all carried each others products, either through one site that farms shipping out to the closest distributor or through individual sites?
I can see a consignment system, but with a marginal upfront fee to cover becoming a member of the collective
I'm happy to throw money into that. I produced the dCoreDuino with an inline i2c bus with exactly that in mind. I also think it's good for Arduino, but lets be transparent about it, money will be spent, time will be contributed, money may or may not be made, and at the end of the day, the main beneficiary will be Arduino and the opportunity sales spurned on by a range of i2c devices built with it in mind.
So I think, at the very least, they might provide some guidance about how these collaborative projects might best be organized, and how they might support a range of these devices.
What is the pinout of the i2c port on the dCoreDuino?
And a question to all you Arduino/compatible product manufacturers and resellers - have you ever considered putting all your products into one site, rather than operating individual sites?
Have you ever considered that together it might be possible to build a single e-commerce entity that could rival the likes of Sparkfun?
Interesting... a collaborative e-business. Implementing regional distribution centers make more sense, the difficult part is finding or writing the software to control all the operations. The e-commerce software is still very immature, and there is a huge opportunity for someone who wants to do it right... yahoo couldn't do it right, neither amazon.com...
If you all used the same company to manufacture the PCB's, then you could use them as a drop shipper. I can get 100 boards including PCB design for under USD$200. The multiplexer board is quite big, it was the same price as the dCoreDuino and the designer made a few enhancements to the circuit at no extra charge.
Pinout (there are 3 inline SDA & SCL bus connectors) is A4+A5 with two 1.8k resistors - although, they can be any value, I didn't use SMD's so bench components can be used to test different values. I've read some have used 10k, others less - I tested 1.8k between two Arduino's, and they worked flawlessly.
All true, and I'm sure they're listening so let's see what they have to add.
One thing I'd hate to see happen is us to degenerate into who paid for it, who gets to profit from it.
That wouldn't be difficult, it doesn't cost much to create a private LLC (especially in Wyoming, lol) and contributors become shareholders. That way, it's run like a small business and each persons share is proportional to the amount they've contributed. "Contributors" spending "time" get shares equal to an agreed hourly rate. Of course, the business would run at a loss, but that's not a bad thing if you've got money that would otherwise be spent on coughtaxes
That's true, but who's going to run the site, and would they have to pay for the stock of so many different products or would all the individuals give a supply of their designs as a consignment?
I'm fairly certain NKC was pushing for the opposite idea entirely. Currently, there are small-time sellers spread out everywhere, what if they all carried each others products, either through one site that farms shipping out to the closest distributor or through individual sites?
I can see a consignment system, but with a marginal upfront fee to cover becoming a member of the collective
I know he was, but I can also see people suggesting one site to take orders but everyone fills their own orders. That wouldn't work because of shipping.
He's a private contractor we've hired and I haven't asked him who he uses to manufacture the PCB's.
That is a really good price. I was looking at goldphoenix, it's about $100 for 155 square inches, which should be good for about 20 shield-sized boards.
If there's a collective formed, I'll bring him to the table - it's actually "a lot" under USD$200, and he's located in an area that would be ideal as an international distribution hub.
For a business model, you guys might want to look into ponoko.
Check out how their business model is set up... A derivative of this would be real cool
How about this for a possibility, as local manufacturing would not be practical.
e.g. 100 boards are made. 5 Distribution points all 5 would get 20 boards. Customer orders a board from the most local to him/her.
Designers here too could upload boards that customers could order. Here both the New company and the designers would get a cut.
Of course some of the less run designs would have to be set up similar to batchPCB (sparkfun)... However, a deal could be struck with GoldenPhoenix too (as BatchPCB) did for this kind of service. As designs get more popular, they could be Run ahead of time to save time and money.
Designers here too could upload boards that customers could order. Here both the New company and the designers would get a cut.
Of course some of the less run designs would have to be set up similar to batchPCB (sparkfun)... However, a deal could be struck with GoldenPhoenix too (as BatchPCB) did for this kind of service. As designs get more popular, they could be Run ahead of time to save time and money.
I think we're talking about more than just boards though, we're talking full kits. This means, not only would you need just in time (JIT) PCB manufacturing but also a JIT component supply. You'd have to feel the market out, but I imagine most consumers are looking for 'ships today, tomorrow, or the day after', not 'ships in two to three weeks when all the parts and the pcb come in'. This limits a model like this to very quick-turn PCB lines (and overnight/1 day manufacturing of a single board or panel isn't cheap nor the quick shipping) and very fast suppliers (mouser, digikey, portions of newark, etc). Practically you've just pushed all the startup cost onto each end consumer.
I think realistically, the barrier to becoming a designer is having a (small) run of boards made and purchasing enough hardware to make kits for them. Assuming there is a collective though, these costs could be limited by creating panels instead of single designs and bulking up everyones orders for quantity discounts and access to larger distributors.
The collective model interests me because cheap PCB's were just a part of it. Components, assembly and distribution also happen to be well below anything you'd find on google.
Besides, someone distributing to Eastern Europe might not get very many sales, whereas, someone stateside might run out quite quickly.
So if it's centralized, inventories can be managed more efficiently, that would be the main advantage. But I'd still like to hear from others who are running these small e-commerce sites, and why they haven't already pooled their resources into one operation because there may be issues that aren't immediately obvious, like, maybe some people simply don't like playing with others :-?
"I think we're talking about more than just boards though, ...."
Fair enough, though think of the potential for such a model. Make the boards available this way, as they get more popular then kits would make more sense. Also, the service could provide key components.
True, ideally a kit would be available promptly... however, this could be a good way to get a design off the ground.
Do understand, that I don't mean to duplicate the ponoko model, but rather to adapt it in a way that would make sense for the product.
Pinout (there are 3 inline SDA & SCL bus connectors) is A4+A5 with two 1.8k resistors - although, they can be any value, I didn't use SMD's so bench components can be used to test different values. I've read some have used 10k, others less - I tested 1.8k between two Arduino's, and they worked flawlessly.
Are the 3 inline SDA & SCL bus connectors coming from really different sources or is the same SDA & SCL repeated 3 times? I like the idea of having also +V and GND with each SDA/SCL pair, to be able to have "independent" uShield boards that can be plugged directly to the socket, instead of having to connect wires.
SDA & SCL are repeated 3 times. The 3 instances are located at the bottom right of the board, and at the 'extreme bottom right' are vcc & gnd.
Unfortunately, there was only so much room available and the board could have been quite big had we added everything we wanted.
But the main objective was to keep it to a reasonable size, it's about 10mm wider than an NG and about the same length, but packs the punch of two NG's connected over the inline i2c, minus the USB, that we don't need for the project the boards are intended.
We could have paired vcc & gnd, and had just two instances, we'll make another version after we've had a chance to evaluate these along with the I/O multiplexer boards, so perhaps we'll do that next time around.
but the original creators gets very critized by other "engineers" because there is always a better way of doing things... ... I didn't see such behavior in the software world.
Ummm, you don't? Linux vs BSD, FreeBSD vs OpenBSD, vi vs Emacs, python vs perl, plone vs drupal, etc, etc...
I guess that is why there are different versions of Ethernet shield...
I was intrigued the Ethernet shield developments were mentioned a couple of times.
In that case there's at least one technical reason for the concurrent development in that each of the shields I've seen mentioned have three completely different underlying components used in the implementation. Each of the components have trade-offs in terms of functionality, price and resource usage which different people value in different ways.
And for me at least, I didn't start out to design an Ethernet shield--it just sort of developed from something I started playing with.
I'm observing the conversation with interest as it touchs on questions of product development from prototype I've been wondering about.
I don't, however, happen to think hardware people are any more egotistical than software people but YMMV.
What kind of tools do you use? I use Eagle, mainly because of the free version. The restrictions are lax enough to create a full-sized shield.
Which raises another issue for hardware developers who also rate Free as in Freedom an important attribute--can a design be said to be Free if it requires non-Free tools (even if they are free from financial cost at some level).
Even then, you've got multiple choices of Free tools--I use KiCad but others might prefer gEDA.
What kind of tools do you use? I use Eagle, mainly because of the free version. The restrictions are lax enough to create a full-sized shield.
Which raises another issue for hardware developers who also rate Free as in Freedom an important attribute--can a design be said to be Free if it requires non-Free tools (even if they are free from financial cost at some level).
Even then, you've got multiple choices of Free tools--I use KiCad but others might prefer gEDA.
--Phil.
Interesting "dilemma" Phil, what if the PCB designer is using a commercial application they purchased?
What about schematics that have been drawn on a bit of paper for others to use for free, was the paper and pen something they acquired for free, to be able to call it free in the first place - is that what your meaning?
