Yes, I finally decided to go with the BSD license, since I'd like to enable people to benefit from this, which might mean being able to add new tests without revealing the model answer :)
While building Texts.com, my CTO and I learned a ton by studying the path of GetchaBooks. They had shut down and decided to open source their code (http://getchabooks.com/open-source.html), which is a generous and noble thing to do.
I see it as good coding karma; and you never know the contacts you might make (I initially wanted to hire one of the creators; but I settled for buying him a few beers since he had just accepted another offer).
Hi Frank, I wrote most of the backend for GetchaBooks. Glad you got some value out of the source.
Even though I invested a fair amount of time in continuous refactoring while actively developing it, refactoring and documenting to a point where I wasn't embarrassed to make it public and could realistically say that someone else would be able to use (which ended up happening with tuftstext.com) and perhaps modify or extend it was a significant learning experience.
IIRC I spent a few weeks on that, but I think it was worth it. Instead of setting aside the project with a vague idea of the right way to do some things, I ended up actually having done them the right way, so now when I encounter a similar problem I know how to do it the right way the first time.
This is basically the same situation you would face if you were handing off code when leaving a company. This experience is one reason why I think it's important to develop and document your project to open source quality standards throughout the lifetime of the project.
And what better way to do that than to open source your code even for an active company? Usually you're probably not competing on tech, and I bet any community contribution you'd get is greater than the likelihood that someone will actually manage to maintain an instance of your code.
The books-by-course space is a perfect example of this. So many "startups" have duplicated so much work building these sites, when what you actually need to succeed is marketing. (The thing we took away from it, I think, is that you need per-school branded sites.) They should start sharing resources more.
What do you know, here's the college bookstore part of GetchaBooks extracted into an HTTP API, now with parallel scraping, that I just made public on Github: https://github.com/mwhite/bookstore-api Maybe it wouldn't be too hard to get working. But expect to get sued.
It is very cool to open source your software. That being said, I'd recommend you work with a native English speaker on your marketing if you plan to pursue another startup. The copy writing on the marketing site is bad enough that your code never mattered.
As a techie who is doing his own thing, I can see why jcampbell would assume English was not your native tongue. Copy writing is really an art form where it's almost a language of its own. Although your sentences may be grammatically correct, the nature in which they are presented seems unnatural.
To be honest, even if you involved an English professor, the outcome would probably be the same. Asking a native English speaker for marketing/copy writing advice is not the right thing to do. Find somebody who is good at copy writing or learn how copy writing works.
Even if you are a native speaker, some outside advice would help.
> "Recruit Top Developers with Feedback-Driven Interviews"
That doesn't make logical sense. Recruiting is about finding candidates. "with interviews" makes no sense in this context.
"Feedback Driven Interviews" means nothing to anyone. Google doesn't know what it means. It is not an industry term. It is bullshit bingo. That never works.
Why not start with "Improve your hiring process by ..."?
I am no copy writing pro, but you need to learn to invert. Become the customer and sell yourself. Write in the active voice.
Native english speaker here. I consider CV reviews, phone screens, in person interviews etc to be part of the 'recruiting' process which encompasses basically everything before the candidate signs a contract and becomes an employee.
So 'recruiting with interviews' makes sense to me.
I feel bad your startup failed, but I'm grateful you decided to open source the project. I wish more people did the same since there are plenty of lessons to be learned. Plenty of public bulletins and blog posts discuss -- sometimes at length -- over what part management difficulties and other obstacles have played, but very few actually release technical details. Even more rare to see the source released.
Hope you fare better in your next venture. If you haven't started, you should.
That's the idea. The plan is to move anyone who wants to keep data onto private Azure instances, and then to close up shop. Maybe if anyone is interested in developing a customized version of the code, then I'll work on that in the future.
Can you also share insights on how/why the startup failed? Also, do you plan to keep the site operational for the free customers or completely shut it down?
The plan was just to take down the site completely and redirect the domain to the project page for now, but if there's a demand for the free version, then I could be convinced to keep it up. Right now, I'm not seeing that demand though.
I should be a target customer, since I periodically interview candidates for programming positions.
However, landing page and screenshots failed to explain to me what the unique selling point was.
How many people here know what "Feedback-Driven Interview" is?
Whole product seems to be about providing an online testing tool ala topcoder.com. But coding tasks are usually only a small part of the interview process (not that I agree with that, but that's what happens at 90% of companies I've applied for).
I think you've hit the nail on the head here. For one thing, I didn't put enough energy into any of the non-dev aspects of the project. Also, I started writing the code before I had really validated the product properly. I had some recruiters which were interested, but most eventually turned out to dislike the idea of scaring away their prospects with tests, and didn't see it adding enough value. It was even harder to market to individual companies looking to hire.
Yep, it's easy to spot those mistakes in others', when you look from a side. It's much harder for me to do boring work myself, instead of playing with new frontend libs and writing code.
At least you'll be more self-aware next time, few people have this luxury.
I thought I was the only person who attempted at creating startups with asp.net/azure before I switched to RoR. It looks like I was not alone! Kudos to the OP for open sourcing it.
I actually used MVC on the second part of this project, which is not included here. To be honest, my requirements were so small (and ASP.NET MVC seemed quite heavyweight for my needs) that I just went with the first thing that worked, and webforms were just the path of least resistance.
Btw, could you clarify exactly which type of licence you are using (https://github.com/paf31/initialround/blob/master/LICENSE). Is it something like BSD?