LinkedIn App

hi Everyone,

This is the JSON received from LinkedIn [1].
I have a few queries here :-

1) How should be the mapping of "educations" done?
if it should be "schoolName" : ["Foo", "Bar", "Baz" ], then I plan to use
 the following in @contexts in schema.

=====
"schoolName": {
  "@id": "http://schema.org/EducationalOrganization",
  "@container": "@set"
}
=====

but then, I have to tradeoff the startDate and endDate (which is just yyyy
and not a valid xsd:date) [Also assuming that in some cases, it might not
be available too].
But if this is not the right way and we have to preserve that as well then
what should be done? and how?

2) For the field of "following" where it can be
institutions,companies,people and industries I am using again:-
=====
"followingCompanies": {
  "@id": "http://schema.org/Organization",
  "@container": "@set"
},
"followingPeople": {
  "@id": "http://schema.org/Person",
  "@container": "@set"
}
=====
and so on. Does this seems legit? Or any modifications are required here?

3) What term should be "interests", "positions", "skills" and other fields
coming from JSON should be mapped to in context? I am not able to figure
this out.

Adding to this, as my LinkedIn data is short and crisp, its not exhaustive
for all the fields put in the LinkedIn App, as a result there are many
values which are not being received from my profile. So, as Manu suggested
over the meeting, can you please send me your LinkedIn JSON data, that I
could use for testing?

Note:- I can use @list in future instead of @set, but so far I am presuming
as the data is in order to the best of his/her knowledge but there might be
one case where the use itself has filled it in an unordered fashion. So
considering just that, it uses @set.

I will be using the above information to update the schema and share the
gist link soon.

Regards
~Vikash

[1] - https://gist.github.com/ivikash/6282447