Ask HN: Am I living in a parallel universe to think my idea is super awesome?

14 points by bursurk ↗ HN
So I thought one day while searching for some code snippet on the web (because that's what developers do), how cool it would be to have a website where for a particular problem I can see a list of various implementations in one place. And the code snippets are already validated with test cases and I can try them online to see if it suits my purpose.

I tried searching for such a website. Nothing found.

"I can make this and its an awesome idea" I said to myself.

"I need to be careful, all the traffic I would get would overload my server" the naive self said.

So a few a weeks later I deployed a first version, initially for c#

http://volatileread.com/UtilityLibrary?id=1083

And then I waited..... And waited.. No users.. Shared the link on HN.. Din't even make it to the first page. Opened Analytics, saw 10 sessions. Digged deeper, they were all from me.

Is something wrong with me to still think this is an awesome idea?

20 comments

[ 4.2 ms ] story [ 51.7 ms ] thread
I think you need some serious SEO, because most developers I know just take the first or second applicable result from Google.

See also, StackOverflow. How is your site different from that (except for being more focused)?

Stack overflow doesn't allow you to run solutions nor compare different solutions for the same problem. I know google isn't being nice to me.
Nope, this is an awesome idea, I've had it myself several years ago.

My immediate reaction to your implementation:

- C# is icky.

- Would be cool to automatically compare performance of solutions, e.g. timing, memory footprint, etc.

- "Run" button doesn't appear to provide any output?

When I had this idea, I also wanted to be able to provide sample inputs and outputs, and have all potentially matching functions run, to see if any match.

In general, you've hit the problem of building a community without an actual community magically appearing. Keep at it.

You need to write code to use the object, as per your needs, and then it will show all the Console.WriteLines. performance metrics is something I was implementing before getting demotivated.
Ah, cool! Might be nice to auto-fill an example for that.
Ok, played with this a bit, and added the Hamming Weight snippet. Seems to work fine in the solution-tryer, but I can't add a test case; even the blank default gives:

[CS1501] No overload for method 'TestCase_0710c06f5bfa4a9b9b44678df9dd933e' takes 0 arguments

Again, I'd say keep at it, and try adding a lot more content. Hell, I wonder if this is something you could even hire random CS students to add content to.

Hey.. Thanks for trying it out. I'll see what is the problem with the test cases.. You see nobody was using it..
> - C# is icky.

Your opinion, and adds nothing to the discussion.

It's a great idea, but C# is not popular with the Hacker News audience.
I think the idea is average. If you search on Google, you'll likely find a code snippet for most questions. StackOverflow has the advantage of comments and opinions, which add more value, instead of just a block of code. Right now, you have no content, and the idea looks like an afterthought added to your existing site. If you want to give this a try, then...

1. Get a catchy name, like Snippr.com.

2. Setup categories for different programming languages.

3. Write code snippets for the most common functions people are looking to find. For example, visit Google and type... 'PHP how to', and you'll see frequent search terms, people wondering how to connect to a MySQL database, how to send an e-mail, how to upload a file, how to get the current URL, how to redirect to another page. People are looking for these answers, and they're easy to write. You can write hundreds of these snippets in a day and start building content.

4. SEO. Right now your links are http://volatileread.com/UtilityLibrary?id=1084. They should be http://snippr.com/php/how-to-get-the-current-url.

With that being said, I don't think it'll work. People don't open their browser and visit a site of code snippets. Instead, they have a problem, and they search Google for the answer. Showing up on the first page of Google for these common terms is not an easy task. You're competing with successful sites that have existed for the better part of a decade. I honestly don't think you'll be able to outrank them, and therefore you'll receive little to no organic traffic, and that's the traffic your site will rely on.

My impressions:

Build this into a collaborative desktop app instead. I'd much rather have an incredible UX built on top of code snippets I can drag out into my IDE.

I think this already exists; I've used something similar, but the UX and searching was poor, and their was no collaboration.

You'd have to nail the indexing.

I can't see a real difference to stack overflow.

You should also learn about online marketing. Having a great product isn't enough to get traffic. I don't want to be offensive but your product isn't even very awesome.

Thanks for a honest feedback.. But I think I'll continue working on this.. May be some day I'll have a few contributing users..
I like the idea. Some thoughts:

- Microsoft built this as well, it's called Bing Code Search: https://visualstudiogallery.msdn.microsoft.com/a1166718-a2d9.... It fetches popular snippets from StackOverflow and it should come with nice integration in Visual Studio, but unfortunately it doesn't work for me. I don't know what they're doing with the project but it seems they abandoned it.

- I never heard of your project. I can't use it if I don't know about it.

- It's probably too much of a hassle to search for your site and then search for a snippet. I'm better off just searching for the snippet on Google, much faster. StackOverflow usually provides multiple different ways to solve a problem.

- You don't want people to submit their snippets to your site. Too much effort, won't work. Just get the data from stackoverflow if you're going to pursue this idea.

- How will you make money? Or isn't that your goal?

Thanks for the tips. I would like people to submit their versions of code. The snippets then can be then be compared on the basis of perf & mem (in progress) based on test cases as they share the same contract. But all this would require a community which I don't think the site is capable of.

Making money is secondary, I just need to cover the server costs.

It is an awesome idea. You just have to turn into an awesome product now. Even though the software is there, the product still lacks development (like said here already: SEO, Marketing, etc.)

You need to translate your goals into KPIs and optimize for those.

That being said, I think your idea would be incredibly amazing if it offered some kind of IDE integration. I believe Visual Studio had something similar a few years ago, but I don't believe it ever caught traction. Maybe worth it to research and see why it didn't work, and how you can improve on that.

From my perspective it would be amazing to read discussions about code, and get the most used (upvoted) snippet into my code, without ever leaving my IDE.

Your idea is fine, and looks useful. But as with any idea, the implementation is what matters.