I've been asked to create an application that will aid in finding people that speak a certain language. The idea is to search your database for common, say French, last names and get corresponding addresses. Generally the search would be performed with around 150 common surnames of a specific nationality.
Any ideas on how I could do this without quickly exhausting the API request limit?
I had thought of caching results for a wide geographic area, like the state of Georgia, and then doing narrower searches by city, zip, etc. that just hit my local DB instead of having to make an API call. But then I read that find_person only returns up to 10 results at a time. The caching could still be done, but it wouldn't be as easy or effective in avoiding the API limit. Any ideas?
I'm trying to understand your problem and think I might be missing something. 1,500 queries/day, with 10 results/query (which is much more likely if you search larger areas like states), gets you 15,000 results (or 100/surname).
Do you need more in a single day to meet the goal, or is it picking which ones to search that's difficult, or...?
So, one problem is...what would you call it...completeness? Lets say someone searches for Spanish speaking people in Florida. I don't think 15,000 results would be a complete result set for that search. Also, how would you use the API calls most efficiently? You couldn't exactly go through each city or zip code in a state searching through the 150 surnames each time. And what if you did, but a given name had more than 10 results? There isn't a way to "page" through results by supplying an offset is there?
Another problem I see is if (with a cold cache) someone searches for Spanish speakers in FL, and then Italian speakers in NY. The API limit would have already been reached.
I did misunderstand: I thought you were looking for (as an example) a representative set of Spanish speakers, not all Spanish speakers.
This is a case which we don't support through the free API - it's connected to our restriction to use the free API for consumer-facing applications (v. business-facing applications). For similar reasons, we don't support paging.
We do have a paid product at www.w3data.com that provides a service that could do this - the APIs aren't quite the same, but much of what you want is available. If we were to add a for-pay layer to this API, we would enable these kinds of features and usage.
Hello,
I've been asked to create an application that will aid in finding people that speak a certain language. The idea is to search your database for common, say French, last names and get corresponding addresses. Generally the search would be performed with around 150 common surnames of a specific nationality.
Any ideas on how I could do this without quickly exhausting the API request limit?
I had thought of caching results for a wide geographic area, like the state of Georgia, and then doing narrower searches by city, zip, etc. that just hit my local DB instead of having to make an API call. But then I read that find_person only returns up to 10 results at a time. The caching could still be done, but it wouldn't be as easy or effective in avoiding the API limit. Any ideas?
Message edited by Sean Murphy 5 years ago
Scott_WhitePages – 5 years ago
Hey sgmurphy -
I'm trying to understand your problem and think I might be missing something. 1,500 queries/day, with 10 results/query (which is much more likely if you search larger areas like states), gets you 15,000 results (or 100/surname).
Do you need more in a single day to meet the goal, or is it picking which ones to search that's difficult, or...?
Sean Murphy – 5 years ago
Hi Scott,
Thanks for the reply.
So, one problem is...what would you call it...completeness? Lets say someone searches for Spanish speaking people in Florida. I don't think 15,000 results would be a complete result set for that search. Also, how would you use the API calls most efficiently? You couldn't exactly go through each city or zip code in a state searching through the 150 surnames each time. And what if you did, but a given name had more than 10 results? There isn't a way to "page" through results by supplying an offset is there?
Another problem I see is if (with a cold cache) someone searches for Spanish speakers in FL, and then Italian speakers in NY. The API limit would have already been reached.
Any ideas would be excellent.
-Sean
Scott_WhitePages – 5 years ago
Sorry, I thought I had responded. :)
I did misunderstand: I thought you were looking for (as an example) a representative set of Spanish speakers, not all Spanish speakers.
This is a case which we don't support through the free API - it's connected to our restriction to use the free API for consumer-facing applications (v. business-facing applications). For similar reasons, we don't support paging.
We do have a paid product at www.w3data.com that provides a service that could do this - the APIs aren't quite the same, but much of what you want is available. If we were to add a for-pay layer to this API, we would enable these kinds of features and usage.
Sorry I don't have a better answer!