Apple and MS Battle For Desktop Search Supremacy 707
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."
Uhh, GOOGLE? (Score:3, Interesting)
Google Desktop Search?
Re:They both suck (Score:1, Interesting)
WinFS (Score:3, Interesting)
But instead, they are going to make a background process that just indexes things like Spotlight.
I hope it is at least as flexible as Spotlight, to allow developers to make plugins for their indexing engine so that new filetypes can expose information to be searched.
I also hope they do a good job at making it transparent. I don't want my computer to be noticeably bogged down while it indexes a 4GB movie file (hopefully it won't index it in the first place!)
Re:They both suck (Score:5, Interesting)
Smart folders WILL change the way you use your computer. There's no need to hunt through folders for a certain document, as all organization can be done at a smart folder level. Plainly put, it doesn't MATTER where your data is stored in the file structure, smart folders will allow you to organize everything easily and quickly. Just like file systems make it where you don't care where the bits lie on the disk, smart folders will make it where you don't care where the files lie in the directory structure. This is a BIG improvement.
Of course, you didn't actually bother to think about the point you were attempting to make, because you were rushing to get your post near the beginning of the dicsussion so it could be modded up.
Why am I not suprised (Score:2, Interesting)
Re:Why a desktop search tool? (Score:3, Interesting)
Well, THIS sure is helpful... (Score:5, Interesting)
Duh (Score:2, Interesting)
Of course they will. They have 3+ years to respond to Apple's feature.
I mean, thats ridiculous. Thats like saying "Yeah, this new game has good graphics, but this other game coming out in 2008 will look much better!"
Of course it will. You have three years of additional programming and hardware improvements.
And the winner is (Score:5, Interesting)
not to mention the other companys that have since been making products of this nature
Re:Hmm... (Score:3, Interesting)
Re:Dunno... (Score:5, Interesting)
But now, the promise of these tools - in theory - is that we can eliminate this investment of time. We can drop file wherever we want to, and the searching is instantaneous, by whatever bit of criteria we happen to need, conceive, or have access to, at the time of search.
It's not perfect, though: I know that my sense of organization has devolved since I started using Quicksilver [blacktree.com], and that is sometimes a problem, when I am forced to go manually through folders. Heh, who knows - maybe Apple will release some sort of Spotlight -> Automator transition that allows people to use spotlight queries to actually reorganize their data permanently, not smart folder this and query that, but actually reorganize data in the filesystem based on certain things (kind of like how iTunes manages the folders in its library folder.)
Magic Icons. (Score:4, Interesting)
So what does this really mean? Apple already does this but Microsoft promises to NOT ONLY do exactly the same, but have improved uppon the ideer by their next release.
We have an OS versus a Proposal. How can it be they declare the proposal the winner? By that time chances are OSX will have evolved just a tad bit. It takes less time to develop a feature already implimenten then it does starting from the bottom. Even if you do have somthing to copycat.
Re:Magellan lives on as X1 (Score:4, Interesting)
About Yahoo! Desktop Search
Re:Dunno... (Score:3, Interesting)
You can do this already with the search tools already built into Windows XP. Just type the name, part of the name, or search by type of file.
I don't see that this new "desktop search" thing is going to do anything other than teach people how to be disorganized. So now you can put any file anywhere you want without even knowing where it physically is on a disk. Big deal. The point is the OS still knows where it is, and what happens when something invariably gets erased either through user error or a system crash? You erase a folder now and you (should) know exactly what's in it. With the system they're talking about, you'd just lose a bunch of random files and you'd be coming across stuff you didn't even know you'd lost years after the fact (you'd only figure it out when you actually searched for those files, and you'd probably wonder why the search function is not coming up with anything).
I think the desktop metaphor not only still works fine, it is also necessary. There is real utility in knowing exactly where your OS thinks a file or folder really is - not just smart-search pseudo-folders that automatically update themselves based on your criteria (a neat idea, but this should be an addition to the desktop metaphor, not a replacement for it).
Tweaks and helpful features are one thing, but there's no need to throw the baby out with the bathwater like MS is talking about here. I sincerely hope there will be an option to just keep using your system the way we always have.
Remember how Windows XP turned off things like file extensions by default? Remember how the first thing you did when you got Windows XP, along with everybody else in the world, was to turn them back on? I feel like that's the sort of thing MS is trying to foist upon us again.
New search functions are fine. But I don't need to learn a completely new way of doing things on the desktop. My desktop - and my PC - works perfectly well as it is.
Re:Lol. Mod me redundant. (Score:1, Interesting)
Maybe not, but you don't need to
mdfind "query" | grep "regex"
best of both worlds.
Also, it should be trivial to write a tool using the Spotlight APIs that can use regex syntax.
No it doesn't (Score:3, Interesting)
This helps a lot because for example on the topic of utilitarianism the ppt files are util1.ppt
Re:impromptu poll (Score:2, Interesting)
Re:They both suck (Score:4, Interesting)
iTunes 5 will get the benefits of the souped-up V100 database, though, so searching will be even faster. (This won't affect you unless you have hundreds of thousands of songs in your library.)
Re:Uhh, GOOGLE? (Score:5, Interesting)
Of course, those that have followed Microsoft's career know that their basic strategy is always promising, if not guaranteeing, that the next version of their applications will be perfect. Amazingly, some users still believe this hogwash.
A few misconceptions.. (Score:3, Interesting)
That said, what WinFS is trying to tackle currently is considerably more ambitious than what Spotlight, MSN Desktop, or Google Desktop Search do. The "someday" WinFS is not a background process that indexes text documents. Not even close. What Apple is delivering is a "search thing". That is _one application_ of WinFS, but by no means the point of doing it.
The comparison of Spotlight to WinFS indicates (understandable) misconceptinos about what WinFS does. That's reasonable since the WinFS story isn't universally clear within MS, much less outside it
Oh - about NTFS fragmentation. I've been trying to fight this good fight internally for a couple weeks (it was bugging me). The NTFS people claim that defragmentation on NTFS isn't strictly necessary, but it can make certain disks MUCH better and makes most disks "somewhat" better. There are some people on the NTFS team that would be happy to tell customers not to bother with defragmenters but old habits die hard. In any case, i presented the case for ffs cylinder groups and made sure the NTFS developers i talked to understood it. It's not news to them, and they dont feel there is a significant difference in the observed fragmentation levels in normal NTFS volumes and normal ffs volumes.
Personally, i never run a defragger on my NTFS volumes so in that sense, its no different than ffs derivatives (i dont worry about fragmentation)
In any case, there is no current WinFS plan in which NTFS goes away - WinFS's filesystem component attacks a different problem space than NTFS, and WinFS (currently, and, afaik) needs NTFS under it anyhow.
Re: Indexing a 4GB Movie - you might be surprised what WinFS does when it finally gets all the way cooked. Whenever that is
Re:Microsoft has delivered in the past. (Score:4, Interesting)
"Mac OS X may be a nice-looking overlay to Unix, but it still leaves much to be desired. For example, networking in an environment where multiple servers are used is decidedly flaky, permissions must be changed to do simple things like adding fonts or nonstandard printers, and administrative access is difficult."
"...the view from the trenches is that Windows will be the way to go until an OS that is as user- and admin-friendly comes around."
And another:
"A couple of years after the release of Win 95, I attended an Apple event celebrating the new features in Mac OS 8.0. As I sat watching this operating system version that offered full-screen wallpaper (a feature of Win 3.1), Internet options (catching up with Win 95), systemwide sound effects (another Win 3.1 feature) and more, I said to the longtime Mac user sitting beside me that this was Apple's attempt to maintain parity with Windows 95."
Re:Uh...OS 8.5 (Score:4, Interesting)
Oh, and it had banner ads.
This was nicely unobtrusive until OS 9, at which point Apple made Sherlock the Find command and replaced the simple, clean interface with the bloated "brushed metal" that we see to this day. Same functionality as previous incarnations with a more OMG TEH INTERNETS!!!! emphasis.
Oh, and it had banner ads. AND it was big and ugly. So I hauled in my "sherlock" from 8.6 and used that with my powerbook until I switched over to OS X.
And I didn't do that until they peeled Sherlock back into a separate app (that I've never launched on this machine) and left a useable Find in its place. Which we didn't have at all in between 8.6 and 10.2.
Re:Uhh, GOOGLE? (Score:5, Interesting)
I've evaluated plenty of SQL filesystems, attempted to implement my own, and with mild success, ran and tested many implementations. Here's what I found out:
SQL sucks as a filesystem. While it's great that SQL can store relations incredibly well, make finding files easier, and provide a good, intellegent backup system, its faults are with the implementation, every time.
It requires the programmer to make two choices; "Do I want to include an entire SQL engine in kernel space, or leave it in user space?" and "Do I want the 'parsing agent' to run in user space or in kernel space?".
To anyone who's ever worked with an SQL engine, the answer to the first question is obvious. If you move the entire SQL engine to kernel space, you're introducing kernel bloat in the size of 40-80 megabytes for the software itself (including caches, sql tables in ram, etc). But if you leave it in userspace, every user has to have their own copy of the software running for them, or your parser agent has to have a kernel hook that basically takes the input from the user accessing the file system, and redirect it to the SQL engine itself.
The "Parsing Agent" as it were, is a piece of code that actually rips apart the files you send to it, grabbing the content's type, and any metadata it can filter out of the file. It can then use two seperate transfers to send the file to one table, and the metadata to another. When searching for a file, it simply queries the metadata, and matches a file index to the files located in the data cache. This is how almost all modern desktop search technologies work (Google Desktop Search, Spotlight, and whatever Windows Longhorn will have).
The existance of a good parsing agent makes an SQL file system virtually irrelevant. I commend them for not wasting their time storing the whole files in an SQL database, but the metadata should be. That way, using a common API, all programs should be able to quickly find files they need to operate, making the file system more amorphous, and less rigid. Hell, if software engineers cared enough, we could get rid of the whole idea of a heirarchial file system now; simply tag incoming files with a UID, and write them to disk, making the "Parsing Agent" keep all of the metadata, and letting it deal with finding and opening files. You could have links on your desktop to commonly used searches "All files Containing the word 'Lyrics'" (a common one used during my tests).
Really, I'd love to see what Apple has in store for Spotlight, but I definitely know that Windows Longhorn is better off without WinFS the way they originally planned it.
Re:Uhh, GOOGLE? (Score:2, Interesting)
Sherlock's "Find by Content" feature -- the one that crawled your files slowly -- was one application of the technology, but V-Twin was used for many other things over the years. I believe both Spotlight and the SearchKit are based on updated versions of this same infrastructure. As for why it didn't catch on in 1998: The old (pre-X) Mac OS didn't support multitasking very well (so indexing was intrusive), and disks were a lot smaller (so people didn't need search as much as they do now).
Re:Microsoft has delivered in the past. (Score:3, Interesting)
No. (Score:3, Interesting)
Everything we ship has to live for at least n years, where n changes depending on what it is. We have to patch it, we have to run regressions against it _forever_. When we come up with something else better, we have to convince developers why this is bad and why they should switch. We never, ever get to remove it without upsetting everyone.
Just throwing out something that kind of solves a few Photos/PIM scenarios means we're introducing new concepts and APIs that we cant unload.. even though we want it to do more and to do it better.
My team for instance is way far out from shipping its product. We've been letting key customers work with our unreleased internal milestone bits. Parts of it are utterly broken. It doesn't do anywhere near what it needs to do. We're just getting feedback to make sure we're on the right track and to get people thinking about what's coming and how it may help what they're trying to acheive.
Even so, the overwhelming feeedback is "just give it to us now". I suppose we could, but it'd be unfinished crap (even more so than some other things we _did_ release
As someone on a team who has no idea when their work will see the light of day - i am at least as frustrated as you are about MS stuff not shipping.
But ultimately, it comes down to shipping the right thing even if it takes longer. The risk you take is that you miss your opportunity - it's obviously a tradeoff. I cannot make those sorts of "soft" decisions, and especially not about the WinFS project as a whole. Guys down in the trenches (even very smart NT kernel guys) don't always see the picture the same way the people at the top do.. or even as their trenchmates do. I don't have (or need to have) undying faith in the abilities of the management above me, but the arguments i've heard for doing things the way they're being done are generally not objectinable. Again - the course of action is not obvious, so you dont have unilateral approval
Incidentally, developers dont like 1 billion APIs per year. They dont like it when we "get something out there" and then abandon it.
We've done that in the past and we'll probably do it in the future, but it really sucks and lots of people hate doing it, up and down the chain.
As an aside, one appealing thing about
Normally I'd figure we'd get a warmer response for trying to do the right thing in the first version
Re:Uhh, GOOGLE? (Score:3, Interesting)
The guy who developed the Be filesystem is the same guy who developed Spotlight at Apple, so yeah, it'll probably be pretty good.
Re:They both suck (Score:3, Interesting)
First, Linux is our closest competitor. It's not a very good competitor, for reasons that should be obvious, but it's our closest. We have no desire to advance that. That's purely a business decision. (I'm not a business guy. I don't have an opinion about this. But it's how things are.)
Second, Linux is utterly impossible to support. An operating system where every nine-year-old can run his own kernel is not an operating system that we have any interest in working with. The whole overriding philosophy behind Apple is that working with your computer should be a good experience. It shouldn't be frustrating or unpleasant. You should never have a point where you don't like your computer. If we shipped our internal Linux ports, they would fail to work properly on two out of three computers out there. We'd be generating bad experiences for our customers. That's not how we do business.
Third, the reason we ported iTunes to Windows was to sell iPods and music. Linux users don't buy iPods or music. This isn't just anecdotal; the market research is overwhelmingly convincing. So there's no motivation to port.
Fourth, the Mac mini is $500, and its targeted specifically toward people who already own one computer cobbled together from parts. It's designed to be a drop-in replacement for an old-fashioned home computer with detached display, the kind all Linux users have. They should buy Mac minis instead. And, in fact, they are. We can't keep 'em on the shelves of our stores. Post-sales polling says that something like one in three Mac mini buyers self-describe as being primarily users of Linux.
Fifth and finally, in every single environment where Linux and Mac are viable alternatives, we're taking down business hand over fist. This is most obvious in post production. Discreet and Avid used to own post. Then Discreet started shipping their products in a Linux version last year. Suddenly customers were faced with a choice of a Linux product or an Apple product. Lots of them, on the strength of the marketing buzz, chose Linux. They're all going back. Bunim-Murray bought fifty seats of Smoke on Linux two years ago. Every one of them has been replaced with Final Cut Pro on G5s now. Our solutions work better.
Bottom line: Linux has the raw potential to compete with us. Windows doesn't, nor vice versa. Windows is so insular that a Mac can't really do the job of a Windows computer. Likewise, it's so insular that a Windows computer can't integrate into an open network like a Mac can. We're changing that a little at a time, but it's really how things are right now.
Linux, on the other hand, has the raw, untapped potential to compete with us. They're ten years behind us; we started working on Mac OS X technologies in the mid-1990s back when there was still a NeXT. Linux basically hasn't changed since. Evolution, yes, but no revolutionary changes. No Quartz, no Open Directory, no Cocoa, hell, not even anything that can compare with the Finder. So we're not worried, not by a long shot, but we recognize that if somebody were to take Linux and dump that stupid license mess and really invest time, money and energy in making it a modern operating system, it could potentially compete with us. So we're not interested in calling attention to it.
So no. No Linux ports.