tbull007

I remember when LIFX launched on KickStarter last year, there was a storm of comment on HN, basically saying that it was a scam and shouldn't be allowed anymore. It did seem to kick off changes to KickStarter's policies which were probably well overdue.

https://news.ycombinator.com/item?id=4537340

Anyway, I defended them at the time (disclaimer, I know two of the founders), it's great to see them do what everyone thought they couldn't and bring a product to launch within all the relevant legislation.

Congrats!

Full disclosure, Karina is my wife talking about my (ex) start-up, Trunk.ly.

Yes, I quit my job (Director at PwC) and took the family on quite a ride, including uprooting them from family and friends to move to the Valley.

I've been incredibly fortunate to have one person who has always believed in me. I don't think we hear from the person behind the founder often enough. I'd love to hear and read more tales from startup wives.

NB The choice of phrase "startup wife" was deliberate - there are many wonderful female founders who also have the man behind the woman. We spoke about it and Karina felt Startup Partner was too boring and confusing. She's happy to be the startup wife and chose those words. Two weeks into her first real blog, she's also learning the value of a catchy title!

I know three of the LiFX guys personally - they're all well connected and respected in the Melbourne startup scene.

Are they naive to the challenges of deploying a hardware project? At least some of them are, but the lead engineer has been in this field for a long time. Are they trying to defraud anyone? Absolutely not.

I hadn't heard of this project until I read this article, but knowing the team involved there's just no way these guys are putting their reputation on the line over something they don't personally believe can be done.

Must have missed the memo where the Hacker News community collectively decided to stop seeking funding and shooting for the stars. Reading the tone of these comments, we should now only tackle the problems that we... sorry, you the commentator... 100% know can be solved.

Nope - this page could have been a MySpace or GeoCities page and they'd still have sold out.

This is awesome. I ran my own server for several months for a bunch of 10 year olds. But it was a pain in the butt. Keeping it up, paying the bills and so forth. Minefold solves the problem brilliantly.

The kids (or their parents) each pay their own way, they can choose to pay (and play) as much, or as little as they want. Dave and Chris have been fantastic and the service is continually improving.

I wish them well, it's solving a real problem for me and every parent I've introduced to the service has had no problems paying for their kids to use it.

The other thing I've noticed is that the kids love cloning servers too, so they can easily find maps they like, clone it then blow it up without needing to be invited.

I discovered the service here a few weeks back and I've been using it ever since, great to see them launch now into the "wild" as well.

I love these posts by Matt - big data exploration and statistics for nothing more than the joy of it. Great stuff! In this one Matt explores terms that frequently occur together like "Darth Vader".

I've at least partially answered my question - it looks like all my old links were brought in as "private". That's why the public API won't return them.

I registered an application and put in an upgrade request, will let you know what happens!

At the interview stage (I'm not involved in the app, but I know Dave).

That's great! Does it "crawl" wikipedia live?

Background: Trunk.ly is a social bookmarking service that is highly automated and powered by full-text search. For example, you connect it to Twitter, Trunk.ly automatically extracts the links you share and crawls them building a search index for you.

We decided to release a connector to Hacker News for Trunk.ly and scratch our own itch.

The Hacker News connector scrapes anything appearing under "saved stories" and adds the source link into Trunk.ly, indexing it so you can always easily find the link again.

A couple of quick notes - we know that this is NOT the most secure solution in the world. To do this, we have to scrape the site on a regular interval as you. So we have to store your username and password. We encrypt this and take other general precautions, but I highly recommend that you use a different password for HN from other sites as an additional precaution.

It would be great if there was some nice "third party" way of accessing that useful content, either via OAuth or even just using some API key or something.

There are some limitations: * we don't back crawl every story you've ever saved as this could be a large number of pages and we don't want to harm the server. We limit it to the last 200 stories total. * we only support the default HN username and password (no clickpass). We can do click pass if enough people want it. * there is no submission date I could see to scrape, so stories will be stored in Trunk.ly as at todays date.

Let us know what you think.

Credit: Partially based on the work of https://github.com/edparcell/HNScraper

Thanks Darmesh. Its pretty much the approach we took, I guess the only thing is that we should have walked away at your point 4. We discussed it, what stopped us in the end was that even though the deal had now changed, it was still a deal we would have happily done.

In retrospect it was a bigger deal than we made of it and we really should have walked away. Rather than focussing on the terms we should have focused on the signals this was sending - something fishy was going on and internally these guys weren't communicating.

Although I'll (perhaps understandably as this is my post / reflection that's been linked) not concede that our idea is bad, I do accept what Paul is saying here. I don't feel there was any duplicity or otherwise at all in the process - it's very open and straight forward.

From my perspective, of course Alex and I would have liked a different outcome, but I don't envy the decision that Paul and the team have to make either.

There's more than one post which has pointed out they don't always get it right. Only time will tell - YC was never a make or break for us.

Our immediate response (after initial disappointment) was "crap, maybe they're right". We've gone back and crunched the numbers and we still think there is a substantial opportunity here, but there's no doubt our forward planning is already being influenced by this outcome. We'll make sure we test these assumptions a lot sooner than we were initially thinking about.

Who wouldn't want the experience that was in that room to listen to you pitch for 10 minutes, and who wouldn't take some of that feedback on board and at least pause for thought?

TL;DR - We've launched a new type of social search with http://trunk.ly, where you can search a group of people (experts if you like) that you follow for the information they share online.

That title is a call back to an article we posted here on HN just before Christmas inviting you to our beta http://news.ycombinator.com/item?id=2014567 The HN community feedback helped push us to throw open the doors and make Trunk.ly live.

Over the last three months, Trunk.ly has continued to grow steadily, attracting new users and we've been thinking a lot about how people use the site and what we can do to significantly improve it.

So introducing the new http://trunk.ly - with improved social search.

Now the site not only has personal utility for you (by collecting links where ever and however you share them across the internet, then indexing them so you have a "personal google" of content you like), but you can now search the people and their links on Trunk.ly too.

Our observation was that one of the things that people do when they publicly share content, is express an interest or expertise. If you're a Python programmer, you probably share lots of Python links, and more importantly, you only share the links that you think are worthy of some attention. So why not enable a search that lets you tap that knowledge?

We did and it's now live - love to hear your thoughts and feedback. Would you use it? What else would be useful as a search experience and content discovery from people you follow?

Thanks!