Beyond being leery of the whole project, I can't really say how well or poorly
it would work. You are, after all, talking about at least one query per
keystroke. On top of that you're going to be doing some tricky stuff in a PHP
(or some such) back-end, to figure out what query to use - again, one
execution per keystroke.
It doesn't matter if it's only one user, if they type fast enough it will
generate many server hits: several per second, I should think.
I wonder how Google handles their search suggestion gimmick - I never gave it
much thought. That's as close as I can come to a similar application, off the
top of my head.
The Infoshop by Global Information Incorporated
195 Farmington Ave.
Farmington, CT 06032
860.674.8796 / FAX: 860.674.8341
>From: Mike Spreitzer [mailto:mspreitz@stripped]
>Sent: Monday, September 21, 2009 4:11 PM
>To: Jerry Schwartz
>Cc: 'Michael Dykman'; mysql@stripped; Mike Spreitzer
>Subject: RE: incremental name search?
>Ah, yes, I forgot to describe the server and the load. Suppose my web app
>and MySQL are done via shared hosting by some common hosting business. I
>do expect multiple people to be using my web app, but generally only one
>(usually zero, sometimes one, maybe occasionally a few) at a time. Is
>this going to fly, in terms of latency for the incremental lookups and
>overall load at the hosting site?
>"Jerry Schwartz" <jschwartz@stripped>
>09/21/09 03:47 PM
>"'Michael Dykman'" <mdykman@stripped>, Mike Spreitzer/Watson/IBM@IBMUS
>RE: incremental name search?
>SoundEx doesn't do much for names, or non-English words for that matter.
>Although you could use AJAX to handle the web part of this, I can't
>being able to handle much of a load. I think you'll beat the system to
>to little avail.
>The Infoshop by Global Information Incorporated
>195 Farmington Ave.
>Farmington, CT 06032
>860.674.8796 / FAX: 860.674.8341
>>From: Michael Dykman [mailto:mdykman@stripped]
>>Sent: Monday, September 21, 2009 12:21 PM
>>To: Mike Spreitzer
>>Subject: Re: incremental name search?
>>Perhaps this could help you out..
>> - michael dykman
>>On Mon, Sep 21, 2009 at 11:51 AM, Mike Spreitzer <mspreitz@stripped>
>>> Suppose I have a table of a few thousand people, with a FirstName field
>>> and a LastName field. Sadly, my people are not so regular. Some names
>>> have three parts (e.g., due to marriage) crammed into the two fields
>>> ("Hillary Rodham Clinton"). Some even have titles ("Dir, gastroent.
>>> dept., Fubar hosp. OurTown") wedged in there. I want to make a web app
>>> that searches this table incrementally as I type into a web page in my
>>> browser. I am thinking I will have to do something like continuously
>>> display the top 10 matches to what I have typed so far. Of course,
>>> am typing I do not know exactly what is in the database. I generally
>>> only some of the parts of the name when I am typing (e.g., I am looking
>>> "Mary Jones" without knowing whether Jones is her maiden name).
>>> I am even typing something that is a spelled a bit wrong ("Schiller"
>>> "Shiller") or variantly ("Lizzie" vs. "Elizabeth"). This seems pretty
>>> from what MySQL can do directly. I know about "LIKE" matching. I know
>>> the wider SQL community has something called "soundex", but I have not
>>> found it in MySQL. I have a hard time imagining what will help me with
>>> variants on a name. I do not see any easy way to find the "top 10"
>>> matches. Am I missing anything that can help me here?
>>> Thanks in advance!
>>> MySQL General Mailing List
>>> For list archives: http://lists.mysql.com/mysql
>>> To unsubscribe: http://lists.mysql.com/mysql?unsub=1
>> - michael dykman
>> - mdykman@stripped
>>Don't worry about people stealing your ideas. If they're any good,
>>you'll have to ram them down their throats!
>> Howard Aiken
>>MySQL General Mailing List
>>For list archives: http://lists.mysql.com/mysql
>>To unsubscribe: http://lists.mysql.com/mysql?unsub=1