Dead simple web crawling (at scale) with Zillabyte
blog.zillabyte.comSo just two hours ago the other post by OP (https://news.ycombinator.com/item?id=8143000) was flagged to death. Presumably because quickly after posting, two eerily generic and positive comments showed up (one from a new account).
That being said, as far as the service goes, some comments:
1. The popup "join our newsletter" after a minute was a little annoying.
2. Without guarantees on capacity and utilization, pricing by "machine hour" is useless to me. I'd prefer something more concrete.
Why hasn't someone built this before?
Maybe I'm missing something. I figured you could do this already using a few mashable API's.
There have been efforts. Looking at the product, I see that the component builds on an open source project.
Doesn't passing data through multiple APIs raise concerns of data polution/cost overhead? At least this uses a flat,pay as you go cost.
useful tool