Perhaps my results with tech have been good because I am an engineer with lots of experience designing and manufacturing hardware/software products. I have seen projects on Kickstarter that have me question whether or not the person fully understands what it takes to take a project from concept or prototype to a finished product. Needless to say, I don't back such projects.
The article gives a sense that making hardware products is hard and expensive. The author is most-definitely correct. A product like this light bulb could easily burn-up above US $100K in regulatory testing (UL, FCC, CE, TUV, etc.). This is particularly true if the designer isn't well versed in this area.
The article looses a little authority in my eyes when the author says things like:
> The heat sink is crucial, especially if you want to put lots of wifi electronics into the bulb.
Which is complete and utter nonsense. WiFi has nothing to do with the requirement for a massive heatsink. If it did, you iPhone would be a huge block of aluminum with fins. No, the heatsink is required because 80% (or more) of the energy going into an LED is converted into heat, not light. This data is from a few years ago, but I suspect it hasn't gotten much better than that. Most of the power you shove into an LED converts to heat that you have to manage, hence the large heat-sink.
I did a project a while ago that managed more than one thousand one Watt LEDs in a very small space. I was personally responsible for the design of the electronics, mechanics, optics and thermal management. Thermal management alone took about six months of FEA simulations and validation through prototypes before converging on a solution. I had to design the thermal management system to handle over one thousand Watts. It was safer to assume that all input energy (and more) was converted into heat in order to have some operating margin for derating the cooling system.
I went on reading the article and thinking that, while I agreed that multi-disciplinary hardware projects are hard, this guy was being a little mean-spirited. After all, he was right in that this project required having an understanding of electronic design, mechanical design, thermal management, optical design and the myriad of manufacturing and regulatory disciplines that come with bringing something like this to market.
The author's incorrect statement about the reason for having a heatsink created a negative reaction in my mind that carried over as I continued reading the article. I knew I was biased. I tried to ignore that and continued reading. Then I came to the point in the article where he told us that the project originator was also behind another project that amounted to not much more than some folded carboard. A project that was supposed to ship in July and, as of four hours ago, hasn't started shipping yet. That was enough to change my mindset. I was, at that point, aligned with the author in sharing his skepticism. If this guy (or guys) can't deliver a cardboard box in nearly three months I can't see any hope of delivering a complex product such as the LED light bulb in six.
Keep in mind that I have always been a believer that small motivated teams can run circles around large companies with lots of money and people. Projects like this don't cost ten million dollars at a large company because that's what it costs to design the product. They cost that much because such organizations burn a lot more cash per unit of time than a small team. They have a lot more people, resources and infrastructure to support. A large R&D organization could burn a million dollars a day whether they stand still or are working on something. In sharp contrast to that, a small focused team working out of a garage or a modest environment could do the same or very similar work for hundreds or thousands of dollars per day, depending on the nature of the project.
This to say that I am not skeptical of their probability for success because the project is difficult. My skepticism comes simply from the fact that they/he have another fully funded project on Kickstarter, one that my teenage kid could probably --with some some help from dad-- execute on in a month, and they have not delivered yet. It's like proposing that you can build a house when you haven't even demonstrated that you can build a tent.
The question here is more one of execution rather than anything else. I'll estimate that if you are really frugal and have to design this from scratch you are looking at a cost of about $500K. This means real design for manufacturing, thermal design, testing, testing, testing (did I say "testing") --which includes environmental testing-- tooling, certifications, procurement, etc.
That means that, at the current funding level of $780K for a little more than 13,000 light bulbs, they have about $20 per bulb left over to manufacture, test, package and ship. Ouch.
Even if my design cost estimate is off by 100% and they can get this ready for manufacturing for $250K, that means that you only have about $40 per bulb to manufacture, test, package and ship. I am not sure that's enough unless you commit to manufacturing tens of thousands of these per month. One thing is for sure, nobody is making a dime on this one.
Remember, business costs and overhead don't stop at manufacturing. There's far more below the tip of that iceberg.
EDIT: I could say a lot more about the nature of bringing such products to market. Here's an important point that I felt warranted an edit: A product like this can burn down your house. It is imperative that its design take no shortcuts. Testing before release must be extensive and thorough. This could take dozens of iterations until a solid and reliable design is converged upon. This, again, takes money and expertise.