Looking at Desktop Search

Source Title:
Seeking a More Intuitive Search Tool
Story Text:

The Washington Post are running a story n the various DTS apps available. They seem to favor Google and MSN above the others, but the point I found most interesting was this:

But these programs also shouldn't exist: Their capabilities should be built into the operating system, something both Microsoft and Apple are working on. Windows users have a while to wait -- Win XP's successor, nicknamed Longhorn, won't ship before the summer of 2006 -- but Apple's Mac OS X Tiger should arrive this spring.

It does seem rather silly that we (you) would even need a third party app to search files...

I dont run Windows, so have not tried any of them - I am about to buy another machine though, and will run XP on it, if only so i can try some things like this and write a more informed post :)

So, if you're using DTS, what are you using and why?



Windows does have a file search, but it is pretty slow, especially when you have large harddrives of files - which it isn;t really built to handle.

So far as I understand it, the DTS simply apply more advanced information processing, so that searching for information is quicker and easier.

Which is great if someone hacks into your computer and wants to find passwords or business documents. :o


Searching on Linux is equally laborious and shite - unless anyone wants to tell be about something other than 'locate'. It does make me very concientious about file naming conventions and keeping an organized home dir though...

Not quite new

Sure beats me why M$ never put more emphasis on optimizing this all-important feature in their OS. Even under XP their generic search sucks all the way - performance, usability, the works.

DTS is nothing new, BTW: Some five or six years back (maybe more - would have to look it up), AltaVista offered a similar tool for free. Our systems were still on NT4 at the time. Search results were actually pretty good but it was such a memory hog we had to switch it off. On our larger systems the regular index runs might well take a full 24 hours or more, with everything else running at about 10% of its regular speed, ugh.

Regarding privacy, of course it's a high risk feature unless it offers military grade encryption, but then you'll have to fiddle with passwords, re-encryption runs, updating etc. again. And automating this sensitive process beats the whole rationale of employing it in the first place: catch someone unaware right at their PC and you won't even have to resort to good old rubber hose decryption ...


Microsoft OS search is almost impossible,
When using Google DTS I was pretty impressed first, but after a while, mailnly when placing and folding files I got reindexing problems - the G desktop search tends to retreive the old path for files.

Besides that and some regular language problems - It's very helpful, way better than XP's start>>search feature.


I never have had a problem with Windows' built in search. It's certainly not because I have few files either! I don't find it abysmally slow - but then I'm not expecting something on my machine to be fast as things are online - different levels of expectation I guess.

I doubt I'll ever have any interest in anyone's DTS. The difference between a few seconds of time and a few milliseconds of time is really of no importance in the overall schema....


unless anyone wants to tell be about something other than 'locate'.

You're running Gentoo but you don't like the non-graphical search?!

grep -r is all you need. ;)


Chris pirillo posts on that same article, it's a good read...

Out of all the Desktop Search tools out there today, Copernic is the most configurable and brand-independent. There is no hit-you-over-the-head agenda with them. I can easily turn off the icon for and integration of their choice for using Alltheweb as the default "Web search." And as far as its deskbar is concerned, I don't use it. Instead, I rely on the far superior DQSD. But why, you might ask, do I choose Copernic over say, Google or MSN? I think browser-based output is a poor choice for local searching. Simple? Yes. Seamless? Yes. Smart? Hell no.


The difference between a few seconds of time and a few milliseconds of time is really of no importance in the overall schema...

Yeah, Desktop Search has been an of-the-moment buzzword for awhile, but how many folks really, really need it? Is DTS actually a neat little technology -- in all of its flavors -- in search of an almost non-existent market?

We sit in front of out machines for who knows how many hours each day (and night), probably among the heaviest PC users out there in terms of hours per day. Now, I'm not the most organized person in the world, but I know I can usually put my finger on a four- or five-year-old document in a matter of a coupla' so folder clicks.

I occasionally have to use Windows' search to find an obscure system or program file, no problem. And the once every blue moon I have to chase down an e-mail even the clunky Outlook search finds it in about five seconds.

So, maybe we should ask: Why DTS in the first place?

It's a mass factor

As a publisher, researcher and developer, I take a somewhat different view. Obviously, it's a question of sheer mass: with over half a million files on our main systems, with several people creating fresh input every single day, it's pure fantasy retrieving docs within a few folder clicks. (Sure wish it weren't!)

I'm using Eudora for my mail client. Having more than half a gig of e-mail data alone to monitor, even in the current EU version the search function still sucks big time. So I'm using EFind (a free third party app) instead which is ultra fast even though it will create its own set of data hiccups ever and a again.

BTW, haven't used Copernic for system search a lot but its summarizer function is really neat. If you need some fast, manually generated q&d webpage summaries, Copernic does a marvellous job and creates some good text content immensely usable for SEO purposes, cloaked or non-cloaked. :-)

The DTS stuff scales?

I'm kind of surprised. I just assumed (duh?) that these apps were targeted to the single-user market. If you can finagle them to work business-wise, then I can see some use.

Think it's some market in between ...

... single user probably not being in need of this stuff in the first place, while really huge data hogs will more often than not have to cook up their own, proprietary solution.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.