Skip to content

Faster searching mechanism?

April 3, 2009

“Seaching in SuperMemo trough items and abstracts is fast, but not as fast as incremental search on FireFox. To test FF3 searching efficacy and efficiency I pasted together 10 books (>600kb text files) from the gutemberg proyect, then when into searching for infor needed, that was faster than a simple search on a 500 item collection. How can I search mi items fast enough?…”

The above text paragraph was ready to be posted, but somehow it didn’t make it trough…

What did I learn in the last couple of days?. I learn about the algorithms involved in searching big text files, or collection of a big number of files, such as SuperMemo database. I don’t agree with the creators of this program that the best way to go is one file archive per item (even one part of the item as in question and answers o more complex items), this in not efficacious, how ever I learned about indexing and fast search mechanism. Then I had my little eureka!

SuperMemo does have and indexing function (maybe it should have a direct button for it to), as a mater of fact I knew it along time ago (but I didn’t put that snip of info into my beloved program). As time is constantly pressing me on to find more efficient ways to search trough my SuperMemo, I almost went for an external search engine (like googles desktop) but is better if you have it all at hand.

I acomplished this indexing by using the rebuild lexicon function. What does it mean, well for me it means that every single word is indexed and SuperMemo searches that index first in a fast algorithm and then searches in the items or extracts not indexed.

This is what I did: rebuild its lexicon, trough the repair option on file menu. When you do this a message pops out about a taking a lot of time, it took less then 15 min to index all my files in the SuperMemo collection.

Afterwards I found that my searching experience no longer took  more then 3 seconds, but was almost inmediate, ok I’m way to obssesed with time, but 3 seconds on 72,000 repetitions, well thas a huge alot of time. (how do you get to these number of repetitions? any on with over a 2 year use of SuperMemo knows it).

I whish search windows would be more alike other programs and have and easy way to specify which words or phrases to search in a single field, less clicking to different windows is really important. How ever know my search if really fast so I don’t find this incovinient so frustraiting as was the slowliness of the my searches with it.

By the way I use my whole collection in virtual RAM drive, why? can get faster any faster then that. But that’s another story on my optimization on using my chip-based hippocampus: “Supermemo”.

Advertisements
No comments yet

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: