Slashdot Mirror
Apple and MS Battle For Desktop Search Supremacy
markmcb writes "As Microsoft and Apple go back and forth about who came up with what idea first, it's been hard to tell who the real innovaters are. Michael Gartenberg and Jim Allchin of Microsoft give some fair opinions on the current desktop search battle. While they do give credit to Apple's iTunes for search inspiration and to Apple being first out of the box in the OS race, they both imply that Microsoft will provide more robust features with the release of Longhorn."
Wasn't Sherlock on 8.5 the first "desktop search" tool? For the Apple/Windows fight, or did it get web intergration with 9? It's been so long ago I forgot.
Actually Apple had a desktop search as part of the Mac OS at least as far back as 1998. I forget what it was called but it came with a bunch of pre-defined search sites and you could download and add plugins from other sites as well. It was part of the OS search feature, though, and not a plug-in to a web browser.
We can currently download Beagle for open source operating systems and desktops, and it's already somewhat usable. It's written in C# and requires Mono, and I think it's one of the killer apps for OSS too. We've also see it ported to Windows so things are getting very interesting here.
:)
So between Spotlight and Longhorn and Google and Beagle, it's not just a 2-way battle
"Backups are for wimps. Real men upload their data to an FTP site and have everyone else mirror it." -- Linus Torvalds
I'm not sure where that's coming from. I can't speak for Longhorn's search features, but Spotlight (in OS X 10.4) will search based on contents, file name, and tons of metadata. You could toss your files onto your hard disk in any random way you want, and it would be equally efficient at searching as if you had organized it in a more human-friendly way.
Ideally, if you can't remember what you called the document, then maybe you can remember a few key words from its contents, the approximate day when you created it, some metadata such as "photo taken at the Mackinaw Bridge" or something like that.
So while this may not be groundbreakingly new, I think that Spotlight really will provide USEFUL features. Based on what I've seen in previews and whatnot, it would be extremely useful to have an always-ready and always-accessible search feature which can handle metadata easily.
Anyone here old enough to remember Lotus Magellan? If any company or product could be considered first in the desktop search category, it would be Magellan. Released in the late 1980s, it indexed every file on your hard drive into Btrees; when you searched for a term, it would narrow the results in realtime with every keystroke -- blazingly fast. Found files were displayed (many looking just like they would in their native program thanks to several file type filters) with the search word highlighted. Truly one of the MS-DOS highlights of the 1980s.
Spotlight is largely an improvement on the ideas he developed with LiveQueries, adding natural language metadata searching to an OS that's pro-actively metadata oriented in the first place.
If anything, everyone else copied BeOS... the real difference is Spotlight is available to the public at the end of the month. With WinFS, who can say? 2007? 2008? 2009?
The open source world can look forward to Spotlight-like functionality once Beagle and inotify mature, the only real drawbacks are that it's currently rather unstable and written in .NET/Mono
Most of the guys who wrote Magellan got back together and released a new desktop search package called X1. http://www.x1.com
It is quite good, and worth looking at, especially if you were a Magellan fan.
First of all, OS X and Mac OS had a superb search FOR ages which works VERY good. Windows search compare to that is a JOKE. Spotlight is just more branded and search more metadata and gives it in more user friendly form.
Basically everything you said here is wrong.
Ever since Panther, we've had a thing called Search Kit. (The technology behind Search Kit goes farther back than that.) Search Kit would index the contents of readable files, meaning plain text, and allow you to search them.
It was slow, it wasn't extensible, and it wasn't modular.
Spotlight is completely different. Spotlight has a content-search component, but it also has a metadata-search component, and both are linked to data through modular pieces of code called importers. Each importer is associated with one or more file types. When a file of a given type changes on disk (is written to, moved or created), the Spotlight import task (mdimport) calls the relevant importer(s) to re-index the file. These importers are very simple and run very fast. Even on old hardware, the overhead of Spotlight indexing isn't noticeable, in large part because it runs at a very low priority.
So Spotlight is really something new. It's ubiquitous and it's modular and it's fast.
Microsoft's search technology looks strikingly similar on paper. Problem is it only exists on paper.
WinFS has been dropped from Longhorn as it won't be ready in time. Well, actually they dropped it and then came up with something else CALLED WinFS which has nothing to do with what you are talking about. The search in Longhorn is an index system just like Spotlight, and everything still runs on NTFS.
Don't expect Microsoft's new file system to be available before 2010. At this point nobody knows what form it will take. WinFS has been kicked around for about a decade now and nothing has come of it, so Microsoft may choose to make incremental improvements to NTFS instead of going the database-driven route.
I have seen the future, and it is inconvenient.
Spotlight indexes every file on your system for which there's a scraping agent (I forget the correct term). And companies can create those agents for their own file formats and tag all sorts of metadata about files in addition to the raw text content.
For example, if your word processor supports a structured title page (i.e. if it knows who the author is, what the title is, etc.), and if there's an agent that understands its file format, you could do a spotlight query that searched specifically for any file where the author was "Anonymous Coward".
More importantly, after the initial indexing pass (where applicable), spotlight doesn' index files nightly like Sherlock. Spotlight knows when you've been sleeping, it knows when you're awake, it knows when you change files a bit, and keeps its index up-to-date. :-)
Comparing Spotlight to Sherlock is a lot like comparing an RSS-enhanced version of Google to the old world-wide-web worm.... It's an entirely different animal altogether.
Check out my sci-fi/humor trilogy at PatriotsBooks.