Facebook has put together some unique features in its search - perhaps some of them are patented? The functions I'm talking about are driven by three different requirements:
- The fact that their database is gigantic and they cannot just join their needs as needed, as you can usually in a single-home business application with less than a million records.
- The expectations of its users depend on other search queries, namely Google, so long-tail searches are performed by adding keywords to the search name of a person, such as “Orlando, Florida” or “Rotary Club”, (or some other identification value, such as the name of the employer).
- The data architecture seems to be shallow, based on the window that we have on it, viewed from the application (of course, it is not small). What I am saying is that, in addition to the so-called “basic information” in a user’s profile, such as gender and current city, most of what a unique profile does is not rigidly tied to logical columns.
Thus, there is a difficulty in the needs associated with the size of the data set, but with the need to deliver the appropriate results to the user community of users who are not savvy in the search, but have their own expectations and training provided by Google.
Given all this (clarification of my question):
a.) What search features are needed for FaceBook, which we need to consider and deploy in our own search applications / machines? . By necessity, I mean, controlled either by the massive size of the data, set or controlled by the expectations of users, as well as the need for organic growth of the site and increase its relationship between its data - I mean users.
b.) / ?
, , - , "" "". Solr . , , .
, , . , , , . " ", , ( , , ..). , , "" Google, , , , . , , , , .
...
- , ? (.. , ..).
- ? " ". , [ ], -, , , " " .
- , , , ? (: , , , , /, - , ).
, , THEN - ""? (: , , "" - MOL, ... MOL)
, . "" , ?
. , , , , , ( , , ).
, , -, , . , " " "" , FaceBook (, URL- ), , , , . [, , ?]
Facebook, . - , , , , - ( "" " ".
- . , - , .
, , , , ?
FaceBook? , , , , ORM ( ORM) - , SQL, DE- .
- , ? ( ) , - .