omarseyal

what's the killer use case for this api?

Yeah, you could say that. Lots of the features are simply writing uris that trigger intents. That said, they're selling passive tags for $3 each. The user who buys that is likely not going to know how that placing a uri on a tag is easy and free...

love sahil and ajay.

re: square -- consider the source.

it makes sense that the square coo would be "anti nfc." the notion of "nfc payments" as the way of the future detracts from the vision he's pitching -- that is "square payments" are the way of the future.

whether nfc ends up being integrated into square or not, and whether or not nfc is useful in the payments vertical is irrelevant to him. square likely wants you to think about square and only square when you think about mobile payments going forward...

My one recommendation would be to have a plan. Don't quit your job or start a startup just to do so - just because you feel like that's what people on HN do. Do it for a reason - an organic solution you've derived form a problem you see in your life is best. And have a plan, as to how you want to attack that solution and problem. While the plan will likely become obsolete in a week, it helps you iterate when you've already thought about where you're headed...

To answer your question directly though; no, there's nothing wrong with quitting your job.

Interesting niche.

One comment on the front page - I think it's great and clean, but I think it undersells what you're doing. The "highlights" that you have (linked form the pale buttons at the bottom) are very compelling arguments to try your startup ... but they're hidden one click in.

is spam morally questionable? or is it just frowned upon by social norms

I've been in this "business" for a short time, and I've met several. Which goes to show you, your personal experience is generally a bad foundation for making sweeping generalizations.

I just consider dev-shops that are female dominated to be Black Swans. They haven't happened, and they are unlikely to exist right now, but they certainly could have occurred if some social factors were different, and they certainly might in the future...

taproot should have a new york branch. they're not the most "hacker-focused" outfit, but i've enjoyed my work with them. (fyi - this is not a teaching organization, it's more of a mechanism for lending your hacker skills to non-profits etc.)

(http://www.taprootfoundation.org/)

I've thought about this before... Wikipedia covers unstructured data, but we need something to be its analog for structured data. Others have pointed out freebase (which kicks ass IMO), but there's also swivel (http://www.swivel.com/). Swivel's correctness model, however, doesn't seem to be the same open idea that freebase is. It seems more to be focused on data "authorities" and "official" data providers as opposed to just an "accuracy of the masses" system that the more open sites rely on.

for anyone who cares here's the tumblr for one of the authors of the article - http://caffeinated.tumblr.com/

Honestly, Black Swan is not Taleb's best book. I would recommend you read Fooled By Randomness first if you have the chance...

http://www.amazon.com/Fooled-Randomness-Hidden-Chance-Market...

not a quote, but i like the poem "if" by rudyard kipling. it kind of describes the sort of evenness and perseverance that is required of an entrepreneur...

http://www.swarthmore.edu/~apreset1/docs/if.html

1. google (gmail / search) 2. localhost:3000 (or some other port) 3. wikipedia

isn't this part of what what open source is for? you have the interest in an idea being completed, but you don't have the resources to tackle it ... or necessarily the expertise ... so you start an open source project, you rally some contributors from groups / companies / schools that might benefit from it (or benefit from contributing it) and off you go ... right?

So the Apple Design Process doesn't involve _external_ usability testing? I guess I kind of already knew that ... since they're so secretive and all...

dojo has this too ... [http://dojotoolkit.org/demos/fisheye-demo]

Ironically, one of the best places for Rails right now, lies in a place you rarely see innovation -- the big-company-internal-portal market.

Most of these sites are built on old J2EE, STRUTS, or worse ASP stacks. They've got complex configurations, and excessively convoluted deployments, both of which contribute long lead times to updates. Rails allows for a new paradigm that quite closely fits the demands of these customers. Sacrifice a bit on the scalability / reliability end, and take advantage of Rails' agility and simple deployment structure.

Because of their light traffic demands and simplistic feature sets (integrations with other internal corporate services are likely just HTTP based) corporate portals are a fantastic target for early Rails adoption. A single decent developer (the sort you can still _hire) should be able to construct and maintain a fairly complex corporate portal -- reducing development investments (whether in-house or outsourced).

It will be interesting to see which "software as a service" / "software consulting" company first adopts Rails as it's status-quo for this sort of consulting...