Re:Duck Duck Go is starting to get coverage
by epi0Bauqu on Wednesday March 24, 2010 @07:46PM
(#1217237)
It's a complicated mashup of my index/crawler, structured crawls/dumps from crowd-source sites, live vertical APIs, and highly modified BOSS (edited, re-ranked, merged, omitted, etc.). The BOSS piece already uses Bing as a backup and could easily use Ask as well.
[ Parent
| on HN ]
Re:Duck Duck Go is starting to get coverage
by dotBen on Wednesday March 24, 2010 @11:55PM
(#1217671)
How have you found Bing VS BOSS.
I have to projects, one kind of a side project and one a main project that would benefit from BOSS integration but I am not confident the service will be around in 12 months.
I've not played with a Bing API. Thoughts?
Re:Duck Duck Go is starting to get coverage
by epi0Bauqu on Thursday March 25, 2010 @10:27AM
(#1218500)
Bing API is equally simple and response times seem similar, so at first glance they seem to be somewhat interchangeable. But I haven't uses Bing as extensively as BOSS yet.
I'd be really interested to know your thoughts as well when you get into it. So please keep me posted.
I'm curious...how long did you spend making the first rev? Also, about how much have you had to invest in the infrastructure for web-scale searching? Are you crawling in the cloud?
Re:Duck Duck Go is starting to get coverage
by epi0Bauqu on Wednesday March 24, 2010 @10:04PM
(#1217495)
About 6 months before I launched something, which was really a MVP, if that. I've invested in infrastructure, but not a ton. I crawl some from the cloud and some from my own machines.
How do you do the crawls without getting banned? Are you using like 80legs or AWS for from cloud crawling?
Re:Duck Duck Go is starting to get coverage
by jacquesm on Wednesday March 24, 2010 @11:40PM
(#1217647)
Would you qualify it as a meta-search engine or as a search engine ?
Re:Duck Duck Go is starting to get coverage
by epi0Bauqu on Thursday March 25, 2010 @08:00AM
(#1218231)
I classify it as a search engine. I'm in to doing whatever produces the best results. I started with my own crawler before BOSS existed and then switched that piece to it (highly modified of course) to concentrate on where the marginal benefit lies. Techcrunch initially classified it as a "hybrid search engine," so maybe that is a better term?