Promoting social memories on online social networks

1. A method comprising:

by a computer server machine, retrieving a previous communication associated with a first user and a second user of an online social network, the previous communication having been made on a date in the past;

by the computer server machine, calculating a distribution-probability score for the previous communication based on a duration of time during which the first user has not had contact with the second user associated with the previous communication, wherein the distribution-probability score reflects a probability that the previous communication will be shared on the online social network by the first user;

by the computer server machine, determining that the distribution-probability score for the previous communication satisfies a threshold; and

by the computer server machine, in response to determining that the distribution-probability score for the previous communication satisfies the threshold, sending the previous communication to a client system of the first user.

2. The method of claim 1 , wherein calculating the distribution-probability score further comprises:

accessing a social graph comprising a plurality of nodes and a plurality of edges connecting the nodes, each of the edges between two of the nodes representing a single degree of separation between the two nodes, the nodes comprising:

a first node corresponding to the first user; and

a plurality of second nodes that each correspond to a second user or a concept,

wherein the second nodes are associated with one or more of the retrieved previous communications; and

determining the distribution-probability score based on social-graph information corresponding to the first node and one or more of the second nodes.

3. The method of claim 2 , wherein the social-graph information comprises an affinity coefficient of the first user with respect to a particular second user or a particular concept corresponding to a particular second node associated with the previous communication, an engagement level of the particular second user corresponding to the particular second node associated with the previous communication, or a number of second users corresponding to second nodes that are associated with the previous communication.

4. The method of claim 1 , wherein the distribution-probability score of the previous communication is based on an image-recognition analysis of an image associated with the previous communication or an analysis of text associated with the previous communication.

5. The method of claim 1 , wherein the distribution-probability score of the previous communication is based on one or more sentiments associated with the previous communication, or one or more social interactions associated with the previous communication.

6. The method of claim 1 , further comprising caching the retrieved previous communications in a data store associated with the computer server machine.

7. The method of claim 1 , further comprising:

retrieving a set of previous communications associated with the first user, the set of previous communications having been made on a date in the past;

calculating a distribution-probability score for each of the previous communications in the set;

determining a ranking of each of the previous communications based on their respective distribution-probability scores; and

wherein, based on the ranking, one or more of the previous communications in the set is selected as having a distribution-probability score satisfying the threshold, and wherein each of the selected previous communications is sent to the client system of the first user.

8. The method of claim 1 , wherein a date on which the previous communication is sent to the client system of the first user is associated with an event that is related to a past event associated with the previous communication.

9. The method of claim 1 , further comprising sending, to the client system of the first user, a share-value indication relating to a probability that one or more second users will engage with the previous communication.

10. The method of claim 9 , wherein the probability that one or more second users will engage with the previous communication comprises a probability that the one or more second users will reshare the previous communication, submit a further communication about the previous communication, a comment, or a reaction-input to the previous communication.

11. The method of claim 1 , wherein calculating the distribution-probability score comprises accessing a scoring schedule, wherein the scoring schedule specifies a time period for calculating the distribution-probability score for the first user.

12. The method of claim 11 , wherein the scoring schedule is based on a number of communications associated with a particular date.

13. One or more computer-readable non-transitory storage media embodying software that is operable when executed to:

retrieve a previous communication associated with a first user and a second user of an online social network, the previous communication having been made on a date in the past;

calculate a distribution-probability score for the previous communication based on a duration of time during which the first user has not had contact with the second user associated with the previous communication, wherein the distribution-probability score reflects a probability that the previous communication will be shared on the online social network by the first user;

determine that the distribution-probability score for the previous communication satisfies a threshold; and

send, in response to determining that the distribution-probability score for the previous communication satisfies the threshold, the previous communication to a client system of the first user.

14. The media of claim 13 , wherein when calculating the distribution-probability score, the software is further operable when executed to:

access a social graph comprising a plurality of nodes and a plurality of edges connecting the nodes, each of the edges between two of the nodes representing a single degree of separation between them, the nodes comprising:

a first node corresponding to the first user; and

a plurality of second nodes that each correspond to a second user or a concept,

wherein the second nodes are associated with one or more of the retrieved previous communications; and

determine the distribution-probability score based on social-graph information corresponding to the first node and one or more of the second nodes.

15. The media of claim 14 , wherein the social-graph information comprises an affinity coefficient of the first user with respect to a particular second user or a particular concept corresponding to a particular second node associated with the previous communication, an engagement level of the particular second user corresponding to the particular second node associated with the previous communication, or a number of second users corresponding to second nodes that are associated with the previous communication.

16. The media of claim 13 , wherein the distribution-probability score of the previous communication is based on an image-recognition analysis of an image associated with the previous communication or an analysis of text associated with the previous communication.

17. The media of claim 13 , wherein the software is further operable when executed to:

retrieve a set of previous communications associated with the first user, the set of previous communications having been made on a date in the past;

calculate a distribution-probability score for each of the previous communications in the set;

determine a ranking of each of the previous communications based on their respective distribution-probability scores; and

wherein, based on the ranking, one or more of the previous communications in the set is selected as having a distribution-probability score satisfying the threshold, and wherein each of the selected previous communications is sent to the client system of the first user.

18. The media of claim 13 , wherein the distribution-probability score of the previous communication is based on one or more sentiments associated with the previous communication, or one or more social interactions associated with the previous communication.

19. The media of claim 13 , wherein the software is further operable when executed to:

send, to the client system of the first user, a share-value indication relating to a probability that one or more second users will engage with the previous communication.

20. A system comprising: one or more processors; and a memory coupled to the processors comprising instructions executable by the processors, the processors being operable when executing the instructions to:

retrieve a previous communication associated with a first user and a second user of an online social network, the previous communication having been made on a date in the past;

calculate a distribution-probability score for the previous communication, based on a duration of time during which the first user has not had contact with the second user associated with the previous communication, wherein the distribution-probability score reflects a probability that the previous communication will be shared on the online social network by the first user;

determine that the distribution-probability score for the previous communication satisfies a threshold; and

send, in response to determining that the distribution-probability score for the previous communication satisfies the threshold, the previous communication to a client system of the first user.