locksley

Thanks! Similarly, I've convinced my grandma to use this.

Ebay's API might have a 'condition' property, I shall check.

In the meantime it's probably just as effective to put the condition as a query, i.e "iPhone 5 used"

Hmmm, this is difficult to fix because it relies on Ebay's search engine. The only solution I could think of is to add a 'price filter slider' on the results page. This way, you could set the limits to how much you think Dexter Season 6 should cost.

Hey thanks! I'm just using a filter of 2 standard deviations within the median at the moment. Any ideas on how to do it more effectively?

Ah balls. I'm on it.

Hey you sure about that? Seems to be fine for me. But there are two obvious clusters. Perhaps two different LCD models?

http://www.getmarketprice.com/results/thunderbolt-LCD/

This was crazy, thanks for letting me know!

My servers surprisingly seemed to handle it ok with one minor crash.

You don't like the charts?

I'm guess a stats nerd so I like to see that stuff haha.

I like your feature suggestion, just put it in my product backlog.

That's a good idea. I'm planning to build a pricing filter slider on the results page because it seems most people have a rough idea how much an item is worth.

Your problem was the reason why I filtered out results below $100. There's just too much junk below that range like "Fender Bass HELLO KITTY STRAP". Eventually I'll have to think of a way to algorithmically exclude those results.

Thanks for the suggestion.

Yeah, that's on the product backlog! But I'm a full time law student doing this as a side project in order to learn how to code.

NEED MOAR ENGINEERS

Thanks for the comments!

I actually originally intended this for the buyer! I wanted an iphone for myself, so I scraped the 'completed listings' page on ebay, and converted it to csv to build a histogram in excel.

The data is up to the minute. But I cache the queries in order the save computational resources. Each cache expires within 15 minutes.

Initially the data was scraped, but then I moved to the API which improved performance by 200 - 300%.

Ahh, thanks for the heads up. I'm using a URI Escape function on the search terms and that might be breaking something. Looking into it now.

Thanks!

Yeah, I've come across Pricenomics but found their data inconsistent and the sample size too small with ebay results. They're doing good work nonetheless and their front end rapes mine haha. I ain't much of a designer.

Wow, had no idea they could be making so much. I've come across their "Research API" before but didn't think too much about it. Will have a look into it now. Thanks man

At the moment I'm filtering out items 2 standard deviations out of the median. It catches the ridiculous cases, i.e. when some fool tries to get away with selling an iphone for $6000 (yes I've seen this before).

Perhaps I need to filter it within 1 or 1.5 stdevs. Will experiment with this.

However, sometimes you can easily see there are two clusters of results. Not sure how to mathematically determine this. Any ideas?

Thanks for the feedback! Really appreciate it

Ah Crap, thanks for letting me know. I'll patch it up tomorrow, it's 12:30am here in Melbourne!

The most useful tip I could give you (if you're building on rails and deploying to heroku) is to use a Unicorn Webserver. It allows your to handle far more concurrent requests.

This article explains it better than I ever will.

http://blog.codeship.io/2012/05/06/Unicorn-on-Heroku.html

I'm running 8 unicorn workers per dyno at the moment and that's still within the 512mb memory limit because I'm not using ActiveRecord.

This was my weekend project. I made it after being frustrated at not knowing what a fair price for an iphone 5 would be on eBay. I posted this a month ago but wasn't ready for HN traffic. Learnt about load testing the hard way and have optimized it since. So, here it is again.

Feel free delve into any questions.

Is this a YC startup?

Yeah sure, it's a scraper. The histogram is then built out of the 200 results retrieved from 'completed listings', minus the ones that weren't sold. The X/Y scatter plot is built from the 200 upcoming results, intended as a way to find good bargains after you know the median/mean price.

Sorry! It should be ok now.

I made this in rails over a couple of weekends after learning how to code late last year. The inspiration came from wanting to buy an iPhone 5 from ebay but not knowing what a fair price should be. After a writing a script to calculate this for myself, I thought it could be useful for others too and abstracted it into its current incarnation.

Any feedback would be fantastic, especially on UX!