Alexey
Powerful search mechanism for finding language partners Hello there! This is my little essay: The Problem There are a lot of really interesting people on italki.com (as well as on some another, similar resources) to talk with. But it's not always easy to start a conversation, because You know nothing about another person You'd like to talk to. The Solution So it would be great if You could search language partners by words, that could describe yours mutual interests (hobby for example) or maybe another aspects that could tell You that this person do for You. As for me, I'd like to search by following words for example: guitar; harmonica; Blues; bluegrass; bicycles; and so on... But I can't! I just could define some filters like "gender", "speaking language", "learning language", "country", "city". It's useful, but not enough! There is lots of websites in the internet where You could search for language partners. But most of them don't give you an opportunity to define searching filters by words from describing section of user's profile. Why? Conjectural reason for this situation I think it's a technical issue. It's not easy to build a mechanism that could do really fast searching in the long text fields contained several separate words. It's much more easy to search in the fields that contain only one word, for example, searching by a city. You could type "New York" and a database will search by word "New York" in the field "City" of the appointed table. In this case, usually, databases use special mechanism, called "index". It's similar to indexes in books. All records in the table are sorted, so, for database, it's easy to find required record. But when you want to search "New York" in the field that may contains phrase "...I live in wonderful city called New Your...", sorting by the whole value of the field, doesn't make sense! It's because You cannot match values by their whole length. Searching by separate words still possible but whether it will take very long time to search or it will need to build complex mechanism to make search faster.
Jul 25, 2014 8:09 PM
Corrections · 2
1

Powerful search mechanism for finding language partners

Hello there!

This is my little essay:

The Problem

There are a lot of really interesting people on italki.com (as well as on some other, similar resources) to talk with. But it's not always easy to start a conversation,
because you know nothing about the other person you'd like to talk to.

The Solution

So it would be great if you could search language partners by words, that could describe your mutual interests (hobby for example) or maybe other aspects that could tell you
more about this person.
As for me, I'd like to search by the following words for example:
guitar;
harmonica;
Blues;
bluegrass;
bicycles;
and so on...

But I can't! I just could define some filters like "gender", "speaking language", "learning language", "country", "city". It's useful, but not enough!

There are lots of websites where you can search for language partners.
But most of them don't give you an opportunity to define searching filters by words from describing section of user's profile. Why?

Possible reason for this situation

I think it's a technical issue.
It's not easy to build a mechanism that could do really fast searching in long text fields containing several separate words.
It's much more easy to search in the fields that contain only one word, for example, searching by a city. You could type "New York" and a database will search by word "New York"
in the field "City" of the appointed table. In this case, usually, databases use a special mechanism, called an "index". It's similar to indexes in books. All records in the
table are sorted, so, for database, it's easy to find the required record.

But when you want to search "New York" in the field that may contains phrase "...I live in a wonderful city called New York...", sorting by the whole value of the field, doesn't
make sense! It's because You cannot match values by their whole length.

Searching by separate words is still possible but it may take a very long time to search, so the answer may lie in building a complex mechanism to make the search faster.

July 25, 2014
Want to progress faster?
Join this learning community and try out free exercises!