MSN Desktop Search — my take… and IFilter!
I’ve downloaded it, i’ve used it.
You know what i’ve been impressed about it the most?
The ability to extend beyond what can and cannot be searched.
Yeah, ok, it’s fast. But that’s what I expected from it in the first place.
Now, this ability to extend the capabilities of what MSN Desktop Search can, and cannot index, now that’s powerful stuff!
Out of the box, does it search through the contents of a PDF file? Nope.
Out of the box, does it search through the contents of a ZIP file? Nope.
Is there a way to make it search through the contents of a PDF file? Yep.
Is there a way to make it search through the contents of a ZIP file? Yep.
So, let’s look at how we can turn those two “Nope” into two “Yep”.
Well, there is something called an IFilter.
The PDF IFilter allows the Microsoft Indexing Service the ability to index text within the PDF file (which, by default, it won’t do).
Here’s an MSDN doc on the IFilter.
Many Microsoft products out there actually utilise a indexing service, all of which allows for extensibility through IFilters.
An IFilter will “assist” the indexing service to index other file types, such as archives (eg: ZIP/RAR/TAR, etc…) and PDF documents.
Adobe is one company which provides a free PDF reader we all probably use, called Adobe Acrobat Reader. They also provide a free IFilter for the indexing service to allow the service to search through the PDF files. There is a link to the IFilter listed here (under the heading “Searching PDF files”).
I’m sure most of my readers (maybe not all) are programmers, or at least have access to one (for emergency purposes!
). What do programmers like to do? Well, we like to create our own thing when we can’t find anything that does what we want.
So, let’s think about what other types of files that MSN Desktop Search should ideally index, but doesn’t.
Come on, seriously, if you are a programmer, and you come up with a bunch of IFilters that you think would make MSN Desktop Search useful, then yeah, go ahead develop them. And deploy them online. And you know what? Get Robert Scoble to notice them.
I remember seeing a comment on the msn search blog about searching through the email stored in Mozilla Thunderbird and other mail clients.
My response? Ok, I didn’t post it at the time. Create an IFilter that will allow the indexing service to index the mail archives!
Mozilla products are open source right? Well, that means someone out there can easily create an IFilter which allows the Microsoft Indexing Service to “peek” into the file(s) and index the text within.
Here’s an IFilter for Visio 2003.
During my searching I found an incomplete IFilter sample using .NET. And I also found something interesting on PInvoke.NET about the IFilter.
There’s a bunch of IFilters listed here. SharePoint uses IFilters as well, so that’s another plus.
Here is a 30 day eval IFilter for ZIP files called ZFilter.
I’m certain there will be many IFilters available before too long as programmers start to realise they can extend MSN Desktop Search quite easily and for free (if they have time).
I’ve only been able to find IFilters for ZIP files that are available at a cost. Which most end users won’t want, especially since they just downloaded a free search tool.
There’s a bunch of IFilters for sale here.
I’ve decided to temporarily hijack the Channel9 wiki so that people can add free IFilters that they either find, or develop. (And, you can also add other tips and tricks for MSN Desktop Search there.).
Wikis are there to be used, so i’m making use of it!
December 20th, 2004 at 01:38 am
oh god.. i didnt understand one word of that
December 20th, 2004 at 08:58 pm
Wendy, that’s ok!
This is mostly a tech blog