The problem I have with Octopart and other "better search" engines (like Hipmunk, Padmapper, et al) is that the coverage is not 100%--not every item in inventory is shown on the search engine site.
The lack of complete coverage kills me when I'm looking for a very specific edge case, which are exactly the sorts of cases that would most benefit from a better search experience. When I really want that one 10uF/6.3V X7R ceramic capacitor made by TDK, and I can't find it on Octopart, it renders the whole of Octopart useless.
Until these sorts of frontend sites get better adoption from the people who actually supply the inventory (e.g. Octopart -> Digikey, Hipmunk -> Expedia, Padmapper -> Craigslist), they cease to be useful in all the cases where they would provide major benefit.
As for the part you mentioned, I didn't find a capacitor with those specs from TDK, but there are 12 alternatives here: http://bit.ly/hWnqhP
If you know of a supplier of the TDK part, let me know and I'll contact them.
My point is just that in the course of normal use I want each one of these searches, be it for a capacitor or an apartment or a flight, to be very thorough. If I can't guarantee to myself that the search will cover 100% of the available inventory. I can guarantee that if I skip the nice frontend and go directly to the supplier.
FWIW, I picked the wrong one to pick on--Octopart has way better coverage than Hipmunk, Padmapper, or a multitude of other sites.
Mind sending email to eric@padmapper.com about any obvious omissions?
Edit: you can also uncheck the "accurately located" filter and it should show more of the missing Craigslist posts.
I do think Padmapper is great, though. It's still a huge improvement over the Craigslist interface.
I have no idea where you're finding parts at, that they don't have in their database, besides the sleezy feeling parts brokers that tell you they have everything, even if they don't, so you'll contact them and they can try to locate the part for you at $5.00 per IC for a $0.35 part. I have no interest in these people. If I have to resort to them, than I won't be able to find production quantities anyway.
I love seeing successful YC startups that are outside of the fairly insular "startups serving startups and other stuff startup people are interested in" space.
There's so many verticals out there that have passionate users that are being under-served by today's technology. It might be less sexy from a TechCrunch perspective, but I have the sense that's where a good amount of the lucrative business opportunities are going to be for this next era.
We don't need "a better Twitter", "the next Google", or to be "as big as Facebook". But we do need to bring the power of these products to more and more people and markets.
But I agree - from my experience these are absolutely the best avenue for a return on investment. On the flip side, 9 times out of 10, you won't need any outside investment to get the product off the ground, all you have to do is post on a few specific forums and you will have a business overnight.
From a lifestyle perspective, having a wide range of interests and making friends in different fields might help...
Is this mostly a social issue , i.e. sexiness of techcrunch like startups ?
Or is it an information problem, i.e. because it's hard for software engineers to come across worthy problems from other domains , that they have the ability to solve ?
Or is is something else ?
This is arguably one of the advantages of experience: you're more likely to have been exposed to a wider set of problems, many of which are not being addressed (well) by software.
It's not at all a difficult box to break out of, but the easy part is finding an application domain that you can build stuff for. The harder part -- without having experience in that field -- is figuring out how to reach your prospects. Many of these areas have little to no web presence. It's easy to say "just look at a few forums" but that doesn't help if your customer base isn't online at all.
In terms of technology, the first thing we did was actually take a Google-like approach to the part search problem. For example, before we started working on Octopart, you had to use separate search boxes to search part numbers and descriptions. In order to implement search across multiple fields simultaneously we abandoned mysql in favor of a full-text search engine. So, there's probably more overlap with Google search than you would expect looking at our search results.
