Google Desktop Search Beta Release
October 14, 2004 7:31 AM Subscribe
Google Desktop Search Beta Released. A tiny download and the interface we all recognise. As soon as it's finished indexing my email and hard disk I can finally search my desktop hard disk as easily as the internet.
And it also changes www.google.com to include a Desktop option, that invokes the already running local tool. That's clever.
posted by smackfu at 8:03 AM on October 14, 2004
posted by smackfu at 8:03 AM on October 14, 2004
Damn, you're quick.
Google sure did win a Webby Award for Best Practices in 2001, but a lot has changed since then, and the company seems to grow more ubiquitous by the day.
posted by adampsyche at 8:05 AM on October 14, 2004
Google sure did win a Webby Award for Best Practices in 2001, but a lot has changed since then, and the company seems to grow more ubiquitous by the day.
posted by adampsyche at 8:05 AM on October 14, 2004
The ability to record all instant message communication is great...There have been many times where someone has given me a phone number or email address over IM and I've accidentally closed the window before saving. I also like how when I do a normal google search it will include results from my computer, much like it includes a few news results at the top of the results page.
posted by mfbridges at 8:05 AM on October 14, 2004
posted by mfbridges at 8:05 AM on October 14, 2004
You know, I don't really give a flying crap how many times the boyz say they ain't evil. It's just plain a bad idea to let any company have the depth of personal access that people seem willing to happily give up to Google. Or to AskJeeves or Microsoft, before you ask. People who sign up for this and use it are basically traitors to the cause of personal liberty.
And frankly, I find it kind of odd that people find the privacy policy "reassuring". It's so vague that it lets them have exactly what they really want. Your personal details (address, phone number, credit card numbers) aren't useful to them; what bank sites you go to, how often you download porn (and how often you pay for it), etc., -- that stuff is gold, even if they have no official and legally binding clue who you are.
From the EULA:
From the Privacy Policy:
posted by lodurr at 8:07 AM on October 14, 2004
And frankly, I find it kind of odd that people find the privacy policy "reassuring". It's so vague that it lets them have exactly what they really want. Your personal details (address, phone number, credit card numbers) aren't useful to them; what bank sites you go to, how often you download porn (and how often you pay for it), etc., -- that stuff is gold, even if they have no official and legally binding clue who you are.
From the EULA:
Consent to Collect Non-Personal InformationAs I read this, it basically says that you give them the right to collect unspecified but "non-personal" information, such as a log of your activities, possibly including the text of files and specific filenames, but not including your "address." (Please note that "address" would not preclude them from noting your probably location by IP or low-level zip code, though this kind of demographic information is probably not as useful to Google as your activity log.)
Google Desktop Search may collect certain non-personally identifiable information that resides on your computer, including, without limitation, the number of searches you do and the time it takes to see your results. Unless you choose to opt out, either during installation or at any time after installation, non-personal information collected will be sent to Google. This information will be used by Google only for purposes of operating and improving future versions of Google Desktop Search and will not be disclosed to any third party or used for any purpose other than as described in this agreement. To learn more, please read the Privacy Policy located at desktop.google.com/privacypolicy.html.
From the Privacy Policy:
What information does Google receive?I'll say it helps them! With every search that you perform and use, you are kachinging gold dust into their karmic coffers. Use this, and you are giving away something of terrific value to Google, and giving them tools to reach a state where they won't need to buy it from you anymore.
By default, Google Desktop Search collects a limited amount of non-personal information from your computer and sends it to Google. This includes summary information, such as the number of searches you do and the time it takes for you to see your results, and application reports we'll use to make the program better. You can opt out of sending this information during the installation process or from the application preferences at any time.
Personally identifying information, such as your name or address, will not be sent to Google without your explicit permission.
How we use unique application numbers, cookies and related information.
Your copy of Google Desktop Search includes a unique application number. When you install Google Desktop Search, this number and a message indicating whether the installation succeeded is sent back to Google so that we can make the software work better. Additionally, when Google Desktop Search automatically checks to see if a new version is available, the current version number and the unique application number are sent to Google. If you choose to send us non-personal information about your use of Google Desktop Search, the unique application number with this non-personal information also helps us understand how you use Google Desktop Search so that we can make it work better. The unique application number is required for Google Desktop Search to work and cannot be disabled.
Google Desktop Search uses the same cookie as Google.com and other Google services. If you send us non-personal information about your Google Desktop Search use, we may be able to make Google services work better by associating this information with other Google services you use and vice versa. You can opt out of sending such non-personal information to Google during the installation process or from the application preferences at any time.
posted by lodurr at 8:07 AM on October 14, 2004
I think the privacy issues extend beyond the trustworthiness of the company itself. More and more, the government is asserting its 'right' to corporate information about individuals. Look at the credit card industry for example; you might trust Citigroup, but I wouldnt buy all that "fertilizer and fuses" or whatever on the card. Similarly, if your computer is full of porn that you want to search with, it wouldnt surprise me at all if you get denied the security clearance for your next job.
We all draw a line somewhere. I try to buy everything with cash, and I sure as hell am not going to let all my personal files get indexed. Yes, you can "turn off" directories etc from being indexed, but you get my idea.
posted by H. Roark at 8:10 AM on October 14, 2004
We all draw a line somewhere. I try to buy everything with cash, and I sure as hell am not going to let all my personal files get indexed. Yes, you can "turn off" directories etc from being indexed, but you get my idea.
posted by H. Roark at 8:10 AM on October 14, 2004
Hmm. Sounds promising but it's unclear if it indexes mail attachments, archived mail, network folders, acrobat files...? The help is pretty sketchy. I'm currently pretty happy with Lookout for Outlook and Copernic Desktop for files for the time being.
posted by chrispy at 8:10 AM on October 14, 2004
posted by chrispy at 8:10 AM on October 14, 2004
There have been many times where someone has given me a phone number or email address over IM and I've accidentally closed the window before saving.
Your IM keeps this in a log. You get to specify the log. Hell, if you used Trillian, it redisplays the last x lines of a conversation and has a convenient button to open the full log.
I'm perfectly happy using the Find/Search built-in to the system. Once someone figures out how the Google desktop communicates with Google and gives me the settings or the tweak to kill that module, I'll be happy to give the sucker a try.
posted by linux at 8:12 AM on October 14, 2004
Your IM keeps this in a log. You get to specify the log. Hell, if you used Trillian, it redisplays the last x lines of a conversation and has a convenient button to open the full log.
I'm perfectly happy using the Find/Search built-in to the system. Once someone figures out how the Google desktop communicates with Google and gives me the settings or the tweak to kill that module, I'll be happy to give the sucker a try.
posted by linux at 8:12 AM on October 14, 2004
From the FAQ: Since you can easily search information on your computer, you don't need to worry about organizing your files, email, or bookmarks.
Doesn't functionality of this sort lead to a certain laziness in organizing data? I could see casual surfers relying upon this, but anyone with a huge index of files would already have their own organizational scheme. The effort one places in bookmarking or storing files in a directory is an extension of basic problem-solving/memory skills; preceiving it to be a chore only dumbs down one's approch toward reasoning information.
For an idea generator, at least from a user's standpoint, there are far better programs online.
As a means of keeping a leg up over Copernic, WinFS, KDE, or...oh yeah, Finder, Google's very late to the dance. But as noted in earlier posts, this "feature" may really be a market research tool - for their benefit.
posted by Smart Dalek at 8:14 AM on October 14, 2004
Doesn't functionality of this sort lead to a certain laziness in organizing data? I could see casual surfers relying upon this, but anyone with a huge index of files would already have their own organizational scheme. The effort one places in bookmarking or storing files in a directory is an extension of basic problem-solving/memory skills; preceiving it to be a chore only dumbs down one's approch toward reasoning information.
For an idea generator, at least from a user's standpoint, there are far better programs online.
As a means of keeping a leg up over Copernic, WinFS, KDE, or...oh yeah, Finder, Google's very late to the dance. But as noted in earlier posts, this "feature" may really be a market research tool - for their benefit.
posted by Smart Dalek at 8:14 AM on October 14, 2004
A friend of mine installed Lookout on his work machine and the indexing immediately brought his company's exchange server to its knees. Literally, no one in his company could use email.
I'll need serious reassurance that won't happen with the Google tool before I'll install it.
posted by stupidsexyFlanders at 8:16 AM on October 14, 2004
I'll need serious reassurance that won't happen with the Google tool before I'll install it.
posted by stupidsexyFlanders at 8:16 AM on October 14, 2004
Being Wintel only it's useless to me.
I'll have to wait for Google Apartment Search... every once in a while you hear a knock on the door at 3am and it's the Googlebot. You open the door, it pops in and starts looking around your place muttering "Indexing, indexing...". Then you can just go to a web page to find out where you left your socks :-)
posted by clevershark at 8:17 AM on October 14, 2004
I'll have to wait for Google Apartment Search... every once in a while you hear a knock on the door at 3am and it's the Googlebot. You open the door, it pops in and starts looking around your place muttering "Indexing, indexing...". Then you can just go to a web page to find out where you left your socks :-)
posted by clevershark at 8:17 AM on October 14, 2004
M. Roark: It's the potential for government-private cooperation, with the aim of circumventing constitutional protections, that worries me. More and more we see the fed essentially outsourcing to avoid civil liberties issues. Here, for example, they could go to Google, pay a small fee, and discover that someone in the 101-151 block of Pearl Street was sending a lot of emails soliciting people to look at kiddie porn. So they use that as a way to narrow down onto a likely suspect without ever having had to do a "sneak & peek" or get Google to give up anything "personal" (I can think of a couple of ways to do it) -- hell, they could even outsource that to Google.
posted by lodurr at 8:18 AM on October 14, 2004
posted by lodurr at 8:18 AM on October 14, 2004
A friend of mine installed Lookout on his work machine and the indexing immediately brought his company's exchange server to its knees.
Friends don't let friends use Exchange Server :-)
posted by clevershark at 8:18 AM on October 14, 2004
Friends don't let friends use Exchange Server :-)
posted by clevershark at 8:18 AM on October 14, 2004
Use this, and you are giving away something of terrific value to Google,
Well, it's not as if they haven't been practically giving away (with text ads, sure) for free up until now.
posted by adampsyche at 8:19 AM on October 14, 2004
Well, it's not as if they haven't been practically giving away (with text ads, sure) for free up until now.
posted by adampsyche at 8:19 AM on October 14, 2004
Doesn't functionality of this sort lead to a certain laziness in organizing data?
There is a school of thought which regards such laziness as a virtue. I don't claim to understand why humans subscribe to it, but I know why it's good for Google: Because human-specified hierarchies and taxonomies are not as vulnerable to mathematical analysis as are full-text search results.
posted by lodurr at 8:21 AM on October 14, 2004
There is a school of thought which regards such laziness as a virtue. I don't claim to understand why humans subscribe to it, but I know why it's good for Google: Because human-specified hierarchies and taxonomies are not as vulnerable to mathematical analysis as are full-text search results.
posted by lodurr at 8:21 AM on October 14, 2004
Well, it's not as if they haven't been practically giving away (with text ads, sure) for free up until now.
Well, it's not as though that's actually in any way relevant.
I don't make a deal with Google to give them anything in exchange for their search results. Whether they or you think I do, I don't. What happens is that Google makes a bet that they'll get something back from dealing with me.
Now, they can take various technical steps to make it difficult for me to deny them market research data, and that's fine; I'm willing, as little as I use Google, to let them have it. But why in the world should I put myself into a situation where I actually have made a deal with them to let them have the data?
posted by lodurr at 8:24 AM on October 14, 2004
Well, it's not as though that's actually in any way relevant.
I don't make a deal with Google to give them anything in exchange for their search results. Whether they or you think I do, I don't. What happens is that Google makes a bet that they'll get something back from dealing with me.
Now, they can take various technical steps to make it difficult for me to deny them market research data, and that's fine; I'm willing, as little as I use Google, to let them have it. But why in the world should I put myself into a situation where I actually have made a deal with them to let them have the data?
posted by lodurr at 8:24 AM on October 14, 2004
Whether they or you think I do, I don't.
Um, isn't that implicit in the terms of use? As in, if you don't want them to use data, don't use their service?
posted by adampsyche at 8:29 AM on October 14, 2004
Um, isn't that implicit in the terms of use? As in, if you don't want them to use data, don't use their service?
posted by adampsyche at 8:29 AM on October 14, 2004
thanks to the people who suggested copernic. i've been looking for something like this for a while, for my computer at work
posted by lotsofno at 8:36 AM on October 14, 2004
posted by lotsofno at 8:36 AM on October 14, 2004
I want this, but I want the google housebot much, much more. Why are they so anti-Mac, these google types?
posted by bonaldi at 8:39 AM on October 14, 2004
posted by bonaldi at 8:39 AM on October 14, 2004
I like Google, but I won't install this. I don't want my entire hard drive so easily penetrated - unless there's some password feature at the very least, and the index is encrypted up the wazoo. I've known the joy of having Google make everything I have online utterly transparent to anybody. I don't know if I want the same service for my home computer.
posted by scarabic at 8:39 AM on October 14, 2004
posted by scarabic at 8:39 AM on October 14, 2004
People who sign up for this and use it are basically traitors to the cause of personal liberty.
Most ridiculous over-statement I've heard today.
posted by skwm at 8:52 AM on October 14, 2004
Most ridiculous over-statement I've heard today.
posted by skwm at 8:52 AM on October 14, 2004
There is a school of thought which regards such laziness as a virtue.
Indeed... I've known people who put all their documents on the desktop because, as they say, "this way I know where they are".
Then again there are people like me who insist on making lengthy document hierarchies... perhaps it's because I'm a tech writer :-)
posted by clevershark at 8:54 AM on October 14, 2004
Indeed... I've known people who put all their documents on the desktop because, as they say, "this way I know where they are".
Then again there are people like me who insist on making lengthy document hierarchies... perhaps it's because I'm a tech writer :-)
posted by clevershark at 8:54 AM on October 14, 2004
Does this thing do network drives? I can see that it can handle exclusions, but not network shared.
posted by adampsyche at 9:09 AM on October 14, 2004
posted by adampsyche at 9:09 AM on October 14, 2004
bah, since I'm using Thunderbird and OpenOffice this tool is not very useful to me.
But I would like it, it's ridiculous I can search all documents on the web in 2 seconds but looking for a file on my own machine takes several minutes.
posted by sebas at 9:14 AM on October 14, 2004
But I would like it, it's ridiculous I can search all documents on the web in 2 seconds but looking for a file on my own machine takes several minutes.
posted by sebas at 9:14 AM on October 14, 2004
You can opt out of sending this information during the installation process or from the application preferences at any time.
Just turn it off, Lodurr.
posted by F Mackenzie at 9:14 AM on October 14, 2004
Just turn it off, Lodurr.
posted by F Mackenzie at 9:14 AM on October 14, 2004
shares, even.
posted by adampsyche at 9:14 AM on October 14, 2004
posted by adampsyche at 9:14 AM on October 14, 2004
I'm less impressed with an installer that will only install on your System disk and then only if you have 1GB of free space.
Haven't these folks heard of disk partitions?
posted by pixelgeek at 9:22 AM on October 14, 2004
Haven't these folks heard of disk partitions?
posted by pixelgeek at 9:22 AM on October 14, 2004
Then again there are people like me who insist on making lengthy document hierarchies...
I used to be there, and proudly, but then one day, I made a "temp" folder as I was going about some menial task — and then my life went to shambles.
posted by rafter at 9:24 AM on October 14, 2004
I used to be there, and proudly, but then one day, I made a "temp" folder as I was going about some menial task — and then my life went to shambles.
posted by rafter at 9:24 AM on October 14, 2004
On a functionality level, there's one big problem: it only indexes open Outlook files, without including any and all archived .pst files (which contain the material that's usually most relevant when I'm doing deep searching). If and when they address that technical problem it'll be a tool that actually makes my life better.
posted by twsf at 9:31 AM on October 14, 2004
posted by twsf at 9:31 AM on October 14, 2004
Laziness is a virtue. If I wanted to do mental hard work, I'd memorize the Iliad and spread it as a wandering bard. Alternately, I can let my computer remember the Iliad for me (and find it when I've lost it) and spend my valuable mental cycles doing something else.
posted by louie at 9:37 AM on October 14, 2004
posted by louie at 9:37 AM on October 14, 2004
Just turn it off, Lodurr
I won't have to -- I won't be installing it.
My point, if you choose to understand it, is that you techophillic Googleites are choosing to live in a certain kidn of world, and you don't even really understand it. You don't really have an excuse for not understanding it, either. In ten years, when everyone's complaining about the "Evil Google" (or else chanting "I really love Google Brother"), you won't remember a thing about this discussion.
Oh, and, as for the "virtues" of laziness: Haven't you heard? Your mind is just like your body. If you don't use it, you lose it.
posted by lodurr at 9:56 AM on October 14, 2004
I won't have to -- I won't be installing it.
My point, if you choose to understand it, is that you techophillic Googleites are choosing to live in a certain kidn of world, and you don't even really understand it. You don't really have an excuse for not understanding it, either. In ten years, when everyone's complaining about the "Evil Google" (or else chanting "I really love Google Brother"), you won't remember a thing about this discussion.
Oh, and, as for the "virtues" of laziness: Haven't you heard? Your mind is just like your body. If you don't use it, you lose it.
posted by lodurr at 9:56 AM on October 14, 2004
Ok, here's a stupid question: why is this better than the OS search functionality?
posted by sic at 10:54 AM on October 14, 2004
posted by sic at 10:54 AM on October 14, 2004
smackfu: Wow, an actual luddite in the wild.
I would argue that we can use a bit more luddism around. Luddites were not anti-technology for the sake of anti-technology, but concerned about the ways in which mechanization led to industries disposing of much of their skilled workforce. Landauer's The Trouble With Computers points out that the companies that showed the most productivity in comperization in the 1980s, were the telephone companies that replaced large chunks of their workforce with automated switches.
I think some Luddism is well justified in this case. The Google business model is based on mining information from users and finding ways to sell it. I remember not long ago when doubleclick was widely bashed for its invasive data collection strategies. So concern about how and what information it phones home, seems entirely reasonable.
Personally, being more of a unix persion, I've found swish-e to be pretty useful.
posted by KirkJobSluder at 11:01 AM on October 14, 2004
I would argue that we can use a bit more luddism around. Luddites were not anti-technology for the sake of anti-technology, but concerned about the ways in which mechanization led to industries disposing of much of their skilled workforce. Landauer's The Trouble With Computers points out that the companies that showed the most productivity in comperization in the 1980s, were the telephone companies that replaced large chunks of their workforce with automated switches.
I think some Luddism is well justified in this case. The Google business model is based on mining information from users and finding ways to sell it. I remember not long ago when doubleclick was widely bashed for its invasive data collection strategies. So concern about how and what information it phones home, seems entirely reasonable.
Personally, being more of a unix persion, I've found swish-e to be pretty useful.
posted by KirkJobSluder at 11:01 AM on October 14, 2004
Luddites are people who think fire is more efficient than blue screens or core dumps. They have much to learn.
posted by pedantic at 11:01 AM on October 14, 2004
posted by pedantic at 11:01 AM on October 14, 2004
Ok, here's a stupid question: why is this better than the OS search functionality?
Because it's Google, and everything they do should inspire breathless awe in you.
posted by solistrato at 11:06 AM on October 14, 2004
Because it's Google, and everything they do should inspire breathless awe in you.
posted by solistrato at 11:06 AM on October 14, 2004
I'm not giving Google (or any other company) access to my entire hard drive. No way. I thought those 'Gmail is too creepy' guys were nutters, but, well, this desktop search thingy really is too creepy.
posted by reklaw at 11:08 AM on October 14, 2004
posted by reklaw at 11:08 AM on October 14, 2004
To be honest, I don't find the WindowsXP search to be that great, for all of it's friendly dog wizard-like familiarity. It does go much faster if you hunt and peck to turn indexing on. But it's not clear to me whether the box to find text in documents searches for the phrase, or for documents containing both words.
posted by KirkJobSluder at 11:12 AM on October 14, 2004
posted by KirkJobSluder at 11:12 AM on October 14, 2004
I love this. I find it fast and amazing. YEA!
posted by BrodieShadeTree at 11:19 AM on October 14, 2004
posted by BrodieShadeTree at 11:19 AM on October 14, 2004
Did google just eat Mac OS X Tiger's lunch?
posted by putzface_dickman at 11:21 AM on October 14, 2004
posted by putzface_dickman at 11:21 AM on October 14, 2004
Doesn't functionality of this sort lead to a certain laziness in organizing data?
Well, yeah. That's kind of the point. The computer can remember where are your files are, and even infer relationships and organization from their contents. Therefore, axiomatically, it should.
Your mind is just like your body. If you don't use it, you lose it.
And as Dan Quayle said: what a terrible thing it is to lose one's mind.
Being forced to maintain an artificially-imposed organization of documents for the convenience of the computer does not in any way constitute "thinking." In fact, it wastes time that you could otherwise be spending thinking, or creating, or otherwise doing something more useful than shuffling virtual paper around. So you'll actually get more practice using your mind if you let the computer keep track of your documents for you.
posted by kindall at 11:36 AM on October 14, 2004
Well, yeah. That's kind of the point. The computer can remember where are your files are, and even infer relationships and organization from their contents. Therefore, axiomatically, it should.
Your mind is just like your body. If you don't use it, you lose it.
And as Dan Quayle said: what a terrible thing it is to lose one's mind.
Being forced to maintain an artificially-imposed organization of documents for the convenience of the computer does not in any way constitute "thinking." In fact, it wastes time that you could otherwise be spending thinking, or creating, or otherwise doing something more useful than shuffling virtual paper around. So you'll actually get more practice using your mind if you let the computer keep track of your documents for you.
posted by kindall at 11:36 AM on October 14, 2004
I just installed it and it is fast. I think it's faster because it indexes your hard drive instead of performing a live search every single time you do a search. Pretty damned spiffy so far.
posted by ao4047 at 11:43 AM on October 14, 2004
posted by ao4047 at 11:43 AM on October 14, 2004
Ok, here's a stupid question: why is this better than the OS search functionality?
Because it's Google, and everything they do should inspire breathless awe in you.
Are you being intentionally dense? The benefits are speed and depth. Searching my 160-gig hard drive for a file that contains the phrase "my very private important plans for world domination" takes nearly 15 minutes, and probably won't work because I actually typed "my very important private...." How would you like that search to be as fast, accurate, and forgiving as the same search of the web?
I don't understand the backlash here.
posted by gleuschk at 11:48 AM on October 14, 2004
Because it's Google, and everything they do should inspire breathless awe in you.
Are you being intentionally dense? The benefits are speed and depth. Searching my 160-gig hard drive for a file that contains the phrase "my very private important plans for world domination" takes nearly 15 minutes, and probably won't work because I actually typed "my very important private...." How would you like that search to be as fast, accurate, and forgiving as the same search of the web?
I don't understand the backlash here.
posted by gleuschk at 11:48 AM on October 14, 2004
why is this better than the OS search functionality?
I've been at the same job for three years designing stuff, and I frequently have to dig up an old comp to tweak a logo, and that requires searching in many places for *.psd files that might have a certain keyword.
A search like that using XP's native search takes about 5 minutes on a 2Ghz PC that is only a couple weeks old, and free of cruft (but I do have assloads of backups on the drive).
With the Google search, the search takes less than a second.
That's progress.
As to freaking out over the privacy policy, I trust Google won't do anything harmful to themselves, especially if I tick the install option that never sends them any data.
And I should add that I was probably one of a dozen people on earth that absolutely adored Altavista's Desktop Search back in 1997 or so, for the same reason I am already in love with Google's: it searches in seconds what windows takes minutes to do. That's a bazillion-fold increase in response, and worth installing.
posted by mathowie at 12:01 PM on October 14, 2004
I've been at the same job for three years designing stuff, and I frequently have to dig up an old comp to tweak a logo, and that requires searching in many places for *.psd files that might have a certain keyword.
A search like that using XP's native search takes about 5 minutes on a 2Ghz PC that is only a couple weeks old, and free of cruft (but I do have assloads of backups on the drive).
With the Google search, the search takes less than a second.
That's progress.
As to freaking out over the privacy policy, I trust Google won't do anything harmful to themselves, especially if I tick the install option that never sends them any data.
And I should add that I was probably one of a dozen people on earth that absolutely adored Altavista's Desktop Search back in 1997 or so, for the same reason I am already in love with Google's: it searches in seconds what windows takes minutes to do. That's a bazillion-fold increase in response, and worth installing.
posted by mathowie at 12:01 PM on October 14, 2004
No inclusionn of Netscape 7.2, the best browser on the market?
posted by lometogo at 12:07 PM on October 14, 2004
posted by lometogo at 12:07 PM on October 14, 2004
I don't understand the backlash here.
No doubt. It's not like we're giving Microsoft permission to search our drives.......
But then again, *any* data sent back makes things scary. My tivo sends back info, but not enough to scare me away. Same thing with this, not enough info (yet) to bother me. I'm sure the privacy groups will watch this closely - which helps us all.
...but still haven't installed it.
posted by tomplus2 at 12:16 PM on October 14, 2004
No doubt. It's not like we're giving Microsoft permission to search our drives.......
But then again, *any* data sent back makes things scary. My tivo sends back info, but not enough to scare me away. Same thing with this, not enough info (yet) to bother me. I'm sure the privacy groups will watch this closely - which helps us all.
...but still haven't installed it.
posted by tomplus2 at 12:16 PM on October 14, 2004
kindall: Being forced to maintain an artificially-imposed organization of documents for the convenience of the computer does not in any way constitute "thinking." In fact, it wastes time that you could otherwise be spending thinking, or creating, or otherwise doing something more useful than shuffling virtual paper around. So you'll actually get more practice using your mind if you let the computer keep track of your documents for you.
Well....
I think there is a basic fallacy in this line of reasoning that came up on Slashdot recently in proposing to ditch heirarchal file trees with a simple search box. The argument is that people are lazy, so just throwing a search engine on top of the file system will solve all of the problems. However, what advocates of flat file systems have yet to explain is why people who are to lazy to maintain useful file trees, would spend the time entering metadata for every file. Then you run into the other problem. Examining 50 matches for the file that you could be looking for is just as bad as blindly poking around your Documents folder.
As a practical example, I find myself doing a lot of work on Social Network Analysis lately. When I'm searching around for that specific document on Social Network Analysis, I need to know a bunch of additional data about the file:
Is this something I wrote, or a useful paper I downloaded?
If I did write this document, was it for ICLS, AERA, AECT, EIEIO, our local mini homespun conference, a presentation to my co-workers, the draft of my dissertation proposal (which draft?), a draft submitted to a journal (if so, which journal), random notes I wrote after attending a lecture, etc., etc..
Because I've set up a useful indexing scheme for my work, it's really easy to answer what is the source of a document, and what the document was written for. One folder for each project. All files related to that project go in the folder. I have a separate folder for PDF files saved from academic search engines. Dumping files for projects into a folder provides enough critical information that I rarely find myself needing to do a fulltext search (I have a fulltext search set up, I just hardly ever need it.)
Interestingly, Twyla Tharp in The Creative Habit argues that a large part of her so-called creative genius comes from obsessive filing. Each project gets a box, and everything related to the project from cocktail napkin sketches to correspondence goes into the box. She also makes the argument that there is nothing really "magical" to the way she creates art. It is just a matter of organization, hard work, and persistence.
ao4047, mathowie, gleuschk:
I'm wondering what is up with Windows Search. Even with indexing turned on, it seems to go a bit slower than it should.
posted by KirkJobSluder at 12:21 PM on October 14, 2004
Well....
I think there is a basic fallacy in this line of reasoning that came up on Slashdot recently in proposing to ditch heirarchal file trees with a simple search box. The argument is that people are lazy, so just throwing a search engine on top of the file system will solve all of the problems. However, what advocates of flat file systems have yet to explain is why people who are to lazy to maintain useful file trees, would spend the time entering metadata for every file. Then you run into the other problem. Examining 50 matches for the file that you could be looking for is just as bad as blindly poking around your Documents folder.
As a practical example, I find myself doing a lot of work on Social Network Analysis lately. When I'm searching around for that specific document on Social Network Analysis, I need to know a bunch of additional data about the file:
Is this something I wrote, or a useful paper I downloaded?
If I did write this document, was it for ICLS, AERA, AECT, EIEIO, our local mini homespun conference, a presentation to my co-workers, the draft of my dissertation proposal (which draft?), a draft submitted to a journal (if so, which journal), random notes I wrote after attending a lecture, etc., etc..
Because I've set up a useful indexing scheme for my work, it's really easy to answer what is the source of a document, and what the document was written for. One folder for each project. All files related to that project go in the folder. I have a separate folder for PDF files saved from academic search engines. Dumping files for projects into a folder provides enough critical information that I rarely find myself needing to do a fulltext search (I have a fulltext search set up, I just hardly ever need it.)
Interestingly, Twyla Tharp in The Creative Habit argues that a large part of her so-called creative genius comes from obsessive filing. Each project gets a box, and everything related to the project from cocktail napkin sketches to correspondence goes into the box. She also makes the argument that there is nothing really "magical" to the way she creates art. It is just a matter of organization, hard work, and persistence.
ao4047, mathowie, gleuschk:
I'm wondering what is up with Windows Search. Even with indexing turned on, it seems to go a bit slower than it should.
posted by KirkJobSluder at 12:21 PM on October 14, 2004
everything lodurr said. you people are like newly upright simians at the dawn of time. your critical thinking is stopped in its tracks by any old shiny bauble. you'll learn.
posted by quonsar at 12:55 PM on October 14, 2004
posted by quonsar at 12:55 PM on October 14, 2004
No doubt. It's not like we're giving Microsoft permission to search our drives.......
I was about to same the same thing....Microsoft have been searching our drives for ages....and you know, when your porn crashes on your Windows Media Player and you send the report off by mistake....oh dear, your IP and "c:/temp/fuckvid.avi" gets sent to MSoft.
But then again, *any* data sent back makes things scary. My tivo sends back info, but not enough to scare me away. Same thing with this, not enough info (yet) to bother me. I'm sure the privacy groups will watch this closely - which helps us all.
There's an option to NOT sent data back with the Google search.
posted by SpaceCadet at 12:57 PM on October 14, 2004
I was about to same the same thing....Microsoft have been searching our drives for ages....and you know, when your porn crashes on your Windows Media Player and you send the report off by mistake....oh dear, your IP and "c:/temp/fuckvid.avi" gets sent to MSoft.
But then again, *any* data sent back makes things scary. My tivo sends back info, but not enough to scare me away. Same thing with this, not enough info (yet) to bother me. I'm sure the privacy groups will watch this closely - which helps us all.
There's an option to NOT sent data back with the Google search.
posted by SpaceCadet at 12:57 PM on October 14, 2004
Windows Media Player and you send the report off by mistake
you use WMP? you allow it to send reports?
here, put this bandaid on your scraped knuckles, bubbles.
posted by quonsar at 1:06 PM on October 14, 2004
you use WMP? you allow it to send reports?
here, put this bandaid on your scraped knuckles, bubbles.
posted by quonsar at 1:06 PM on October 14, 2004
I take solace in knowing that the GDS is still in beta. I want it to be able to have fulltext searches on my internet history, sure, but far more important to me is searching through my Trillian .log files and my Thunderbird email. Maybe the Trillian logs will be available after GDS has finished indexing. We'll see.
The ability to search network shares, for me, is as important as the Desktop search itself. Too bad this is not yet supported.
I am going to create a Mycroft search plugin for searching with GDS via Firefox. Does GDS always show itself at "http://127.0.0.1:4664/search&s="? Can anyone confirm that port GDS consistently uses port 4664? Email me and let me know, please.
Also, setting the Google Deskbar to Search GDS: very good for ease of use.
posted by quasistoic at 1:08 PM on October 14, 2004
The ability to search network shares, for me, is as important as the Desktop search itself. Too bad this is not yet supported.
I am going to create a Mycroft search plugin for searching with GDS via Firefox. Does GDS always show itself at "http://127.0.0.1:4664/search&s="? Can anyone confirm that port GDS consistently uses port 4664? Email me and let me know, please.
Also, setting the Google Deskbar to Search GDS: very good for ease of use.
posted by quasistoic at 1:08 PM on October 14, 2004
I took off my clothes and stepped into the shower to find another one sitting near the drain. It was about 2 feet tall and made of metal, with bright camera-lens eyes and a few dozen gripping arms. Worse than the Jehovah's Witnesses.
“Hi! I'm from Google. I'm a Googlebot! I will not kill you.”
“I know what you are.”
“I'm indexing your apartment.”
“I don't want you here. Who let you in?”
“I am Google! I find many good things. I find that pair of underwear with the little dice printed all over them. And I watch the tape of you with the life-sized Stallman puppet. These are good unique things. Many keywords and links! My masters will say 'much good job, little robot!' Many searchers will find happy links of Stallman puppet see you! Ahhhh.”
It's coming.
posted by euphorb at 1:34 PM on October 14, 2004 [1 favorite]
“Hi! I'm from Google. I'm a Googlebot! I will not kill you.”
“I know what you are.”
“I'm indexing your apartment.”
“I don't want you here. Who let you in?”
“I am Google! I find many good things. I find that pair of underwear with the little dice printed all over them. And I watch the tape of you with the life-sized Stallman puppet. These are good unique things. Many keywords and links! My masters will say 'much good job, little robot!' Many searchers will find happy links of Stallman puppet see you! Ahhhh.”
It's coming.
posted by euphorb at 1:34 PM on October 14, 2004 [1 favorite]
In Google-space, where are the memory mansions?
KJS: I hadn't heard that Twyla Tharp story, but it resonates with so many things I've seen over the years. My own anecdotal experiences get drawn out and mundane, but I'll point to John Crowley's [obsessively] recurring accounts of Franciscan [order?] "memory-mansions" (see Little, Big and the books in the Aegypt tetralogy). That's what made me really conscious of the "cross-pollination" that was due to putting ideas adjacent to one another. After that, I got in the habit of reading through Factsheet Five cover to cover, just to mingle the juices. It could be amazing fun.
Point being, and as I point out to people all the time, search as a location mechanism tends to constrain serendipity. That's why I loved card catalogs, and hated seeing them go: With a card catalog, you could see things just before and just after your title or see other entries for your target subject. Sure, you could use the digital index that way if you knew how, but most users were never trained to. When I tried to show them, their eyes would glaze over. Once in a while, you'd get a grad student, who really knew the value of serendipity in research, and you'd connect. But most people just wanted a quick answer.
And in any case, as KJS alludes, search is deceptive in that way, too: It's not necessarily quicker, and it's not necessarily saving you any mental effort. In fact, it's costing you mental discipline, and reducing your ability to think through problems in an orderly way.
As for being a "luddite", I won't deny it: I am not ready to accept technological solutions without question. I've made my living by knowing more about technology than 99.49% of the average bears, and in the course of that I've watched it waste so much human effort and mindpower that I can't just sit back and let it go. How we use technology is a choice we make, and it has an impact on not just the world, but on what we ourselves become. And I don't think most people really think much about what that impact is.
posted by lodurr at 1:44 PM on October 14, 2004
KJS: I hadn't heard that Twyla Tharp story, but it resonates with so many things I've seen over the years. My own anecdotal experiences get drawn out and mundane, but I'll point to John Crowley's [obsessively] recurring accounts of Franciscan [order?] "memory-mansions" (see Little, Big and the books in the Aegypt tetralogy). That's what made me really conscious of the "cross-pollination" that was due to putting ideas adjacent to one another. After that, I got in the habit of reading through Factsheet Five cover to cover, just to mingle the juices. It could be amazing fun.
Point being, and as I point out to people all the time, search as a location mechanism tends to constrain serendipity. That's why I loved card catalogs, and hated seeing them go: With a card catalog, you could see things just before and just after your title or see other entries for your target subject. Sure, you could use the digital index that way if you knew how, but most users were never trained to. When I tried to show them, their eyes would glaze over. Once in a while, you'd get a grad student, who really knew the value of serendipity in research, and you'd connect. But most people just wanted a quick answer.
And in any case, as KJS alludes, search is deceptive in that way, too: It's not necessarily quicker, and it's not necessarily saving you any mental effort. In fact, it's costing you mental discipline, and reducing your ability to think through problems in an orderly way.
As for being a "luddite", I won't deny it: I am not ready to accept technological solutions without question. I've made my living by knowing more about technology than 99.49% of the average bears, and in the course of that I've watched it waste so much human effort and mindpower that I can't just sit back and let it go. How we use technology is a choice we make, and it has an impact on not just the world, but on what we ourselves become. And I don't think most people really think much about what that impact is.
posted by lodurr at 1:44 PM on October 14, 2004
im glad quonsar is here to enlighten us simians, just like the monolith in 2001. now if you'll excuse me, i'm going to beat my neighbor to death with the femur of dead critter. my how the condyles pulp his brain matter!
posted by keswick at 1:48 PM on October 14, 2004
posted by keswick at 1:48 PM on October 14, 2004
O'Reilly has an excellent article on it.
"Now, before you start worrying about the results of a local search--or indeed your local files--being sent off to Google, read on. What's actually going on is that the local Google Desktop server is intercepting any Google web searches, passing them on to Google.com in your stead, and running the same search against your computer's local index. It's then intercepting the Web search results as they come back from Google, pasting in local finds, and presenting it to you in your browser as a cohesive whole.
All work involving your local data is done on your computer. Neither your filenames nor your files themselves are ever sent on to Google.com."
posted by quasistoic at 1:55 PM on October 14, 2004
"Now, before you start worrying about the results of a local search--or indeed your local files--being sent off to Google, read on. What's actually going on is that the local Google Desktop server is intercepting any Google web searches, passing them on to Google.com in your stead, and running the same search against your computer's local index. It's then intercepting the Web search results as they come back from Google, pasting in local finds, and presenting it to you in your browser as a cohesive whole.
All work involving your local data is done on your computer. Neither your filenames nor your files themselves are ever sent on to Google.com."
posted by quasistoic at 1:55 PM on October 14, 2004
There may be a way to have your cake and eat it: in addition to opting out, if you're not very trusting but you want to use the software, get ZoneAlarm. The first time it tries to phone home with your seekrit privut infermation, ZA will alert you, and you can tell it right then and there never to allow it. Of course, it'll be sitting there, biding its time, accumulating vital data with which to punish you until one day ZoneAlarm fails to start, Google Desktop seizes its chance and betrays you. Moments later, the black helicopters arrive.
posted by George_Spiggott at 2:15 PM on October 14, 2004
posted by George_Spiggott at 2:15 PM on October 14, 2004
I use Grep for Windows. It works very well. I have a ton of really badly formatted logs from AIM (about 5 years worth I believe), that basically just took whatever came through AIM and put it in an HTML file based on date and persons involved in the conversation. It's really invaluable when you want to know some arcane information about a person in your past and you have 5 years of detailed logs (hey high school, before cars, everyone used AIM). Oh yeah my point is grep is fast an effecient and does not send anything back, innocuous or not.
posted by geoff. at 2:27 PM on October 14, 2004
posted by geoff. at 2:27 PM on October 14, 2004
What really blows my mind is that they launched this without support for Firefox. Okay, I know that (for now) most people use IE, but I'd wager an overwhelming majority of early adopters (ie those most likely to download this) use Firefox.
WHAT.
THE.
FUCK.
GOOGLE?!
posted by keswick at 2:48 PM on October 14, 2004
WHAT.
THE.
FUCK.
GOOGLE?!
posted by keswick at 2:48 PM on October 14, 2004
APPLICATIONS THAT ARE ENTIRELY INCOMPATIBLE WITH DESKTOP SEARCH:
- NOD32 AntiVirus
kthxbye
posted by mr.marx at 2:53 PM on October 14, 2004
- NOD32 AntiVirus
kthxbye
posted by mr.marx at 2:53 PM on October 14, 2004
Are you being intentionally dense?
It's called sarcasm, chumpstick.
And it's because I fundamentally do not trust any company that aggregates that much information, no matter how much geek-happy PR they accumulate or how shiny their logo is.
posted by solistrato at 3:06 PM on October 14, 2004
It's called sarcasm, chumpstick.
And it's because I fundamentally do not trust any company that aggregates that much information, no matter how much geek-happy PR they accumulate or how shiny their logo is.
posted by solistrato at 3:06 PM on October 14, 2004
Won't install on my boot drive because there's less than Gig free (presumable for index files) (and there's no more trimming I can do), doesn't ask me if I want to install elsewhere. Lame, very lame.
posted by stavrosthewonderchicken at 3:37 PM on October 14, 2004
posted by stavrosthewonderchicken at 3:37 PM on October 14, 2004
Beta, very beta.
posted by quasistoic at 4:00 PM on October 14, 2004
posted by quasistoic at 4:00 PM on October 14, 2004
However, what advocates of flat file systems have yet to explain is why people who are to lazy to maintain useful file trees, would spend the time entering metadata for every file.
You can get the computer to do much of that. Computers can make pretty good guesses about document titles, author names, and the like. And machines are already better than humans at distinguishing spam from good e-mail; there is no reason that computers can't create "topic clusters" of related documents using similar techniques.
Creating deep hierarchical organization for our private document collections is something geeks do reflexively, and so they think that it is a fundamental component of thinking, because obviously they are smart and anyone who doesn't think like they do isn't thinking right. So they invent operating systems that force everyone to make their brains work as if they were a geek. People like shallow hierarchies (i.e. documents within project folders, rarely deeper), and they like to keep what they're working on at the moment in plain view. There is no OS that fully supports the way regular people work.
A good litmus test to uncover anal-retentive hyper-geeks: they refuse to let iTunes keep their MP3 files organized for them, because they have painstakingly organized their music files in a way that makes sense to them. Never mind that the file system is a terrible way to browse music; it's always going to be faster to ask iTunes to show you where the file is, no matter how organized you think you are -- assuming you need the actual file for some reason, which you usually don't even if you think you do. Hyper-geeks are big into sunk costs. "I spent all this time creating my own organization for my files, therefore I will further waste my time in increments of one second, because if I don't, I would have wasted the time I spent organizing the files." Given a choice between wasting X units of time, and wasting X+N units time, they will choose to waste X+N units of time "to avoid wasting the X units of time."
Axiom: If what you do can be done just as well by a machine, it is a moral imperative to let the machine do it, so that you can get on with more of the things that make humans valuable.
I fundamentally do not trust any company that aggregates that much information
It's not about trust -- it's about utility. You don't have to trust Google entirely, you just have to find their tool more useful than you distrust them. And it is a very useful tool, and Google has showed itself to be rather trustworthy.
posted by kindall at 4:00 PM on October 14, 2004
You can get the computer to do much of that. Computers can make pretty good guesses about document titles, author names, and the like. And machines are already better than humans at distinguishing spam from good e-mail; there is no reason that computers can't create "topic clusters" of related documents using similar techniques.
Creating deep hierarchical organization for our private document collections is something geeks do reflexively, and so they think that it is a fundamental component of thinking, because obviously they are smart and anyone who doesn't think like they do isn't thinking right. So they invent operating systems that force everyone to make their brains work as if they were a geek. People like shallow hierarchies (i.e. documents within project folders, rarely deeper), and they like to keep what they're working on at the moment in plain view. There is no OS that fully supports the way regular people work.
A good litmus test to uncover anal-retentive hyper-geeks: they refuse to let iTunes keep their MP3 files organized for them, because they have painstakingly organized their music files in a way that makes sense to them. Never mind that the file system is a terrible way to browse music; it's always going to be faster to ask iTunes to show you where the file is, no matter how organized you think you are -- assuming you need the actual file for some reason, which you usually don't even if you think you do. Hyper-geeks are big into sunk costs. "I spent all this time creating my own organization for my files, therefore I will further waste my time in increments of one second, because if I don't, I would have wasted the time I spent organizing the files." Given a choice between wasting X units of time, and wasting X+N units time, they will choose to waste X+N units of time "to avoid wasting the X units of time."
Axiom: If what you do can be done just as well by a machine, it is a moral imperative to let the machine do it, so that you can get on with more of the things that make humans valuable.
I fundamentally do not trust any company that aggregates that much information
It's not about trust -- it's about utility. You don't have to trust Google entirely, you just have to find their tool more useful than you distrust them. And it is a very useful tool, and Google has showed itself to be rather trustworthy.
posted by kindall at 4:00 PM on October 14, 2004
Computers can make pretty good guesses about document titles, author names, and the like. And machines are already better than humans at distinguishing spam from good e-mail; there is no reason that computers can't create "topic clusters" of related documents using similar techniques.
Right. Get back to us on that, mmkay?
posted by IshmaelGraves at 4:33 PM on October 14, 2004
Right. Get back to us on that, mmkay?
posted by IshmaelGraves at 4:33 PM on October 14, 2004
it's always going to be faster to ask iTunes to show you where the file is
I get no small amount of pleasure traversing my deep, anal-retentive directory trees of music files while looking for an album that suits my mood, often serendipitously stumbling across one that I'd rather listen to than the one I had in mind, in much the same way I used to enjoy flipping through my thousands of vinyl albums (the last atoms-vs.-bits way I owned much physical media, thanks to more than a decade of backpacking).
Faster ain't necessarily better. Also iTunes sucks donkey balls on Windows, which I use.
posted by stavrosthewonderchicken at 5:15 PM on October 14, 2004
I get no small amount of pleasure traversing my deep, anal-retentive directory trees of music files while looking for an album that suits my mood, often serendipitously stumbling across one that I'd rather listen to than the one I had in mind, in much the same way I used to enjoy flipping through my thousands of vinyl albums (the last atoms-vs.-bits way I owned much physical media, thanks to more than a decade of backpacking).
Faster ain't necessarily better. Also iTunes sucks donkey balls on Windows, which I use.
posted by stavrosthewonderchicken at 5:15 PM on October 14, 2004
I'm glad I was too lazy to comment in here; kindall said it all better than I could.
stavros, the /. thread said it won't install on a drive with less than a gig of free space (weird that the app's so small but it cancosume half a gig of disk space) and they were killing it for not allowing users to install to a different drive. Hopefully that'll be fixed in the next release, though maybe that's not as easy as I perceive it to be (installing on a drive other than the one being indexed might mess with their setup).
posted by yerfatma at 5:19 PM on October 14, 2004
stavros, the /. thread said it won't install on a drive with less than a gig of free space (weird that the app's so small but it cancosume half a gig of disk space) and they were killing it for not allowing users to install to a different drive. Hopefully that'll be fixed in the next release, though maybe that's not as easy as I perceive it to be (installing on a drive other than the one being indexed might mess with their setup).
posted by yerfatma at 5:19 PM on October 14, 2004
Interesting stuff. I've been obsessed with how to organise electronic documents for a long time, probably because I have a memory-related learning disability. What works best for me is a combination of tagging (meta-data), with search as a back up (advanced/sophisticated: only this file type, only this folder etc). I haven't yet found a way to tag files easily (except for bookmarks, for which I used to use Powermarks until I switched to Firefox) and so I have resorted to folders. For searching files, I have been using windows own.
Over the years I have refined my folder system to create a reasonable level of consistency; normally by thinking hard about and differentiating in various "dimensions" further than by "topic"; such as what the information is to be used for, whether its status will change later, whether it overlaps with other topics, and so on. These days my folder system takes much less time to maintain than it used to, since it has become much more refined. I wish nevertheless that someone would write a Powermarks for files and folders.
Smart Dalek, I really didn't get this: "The effort one places in bookmarking or storing files in a directory is an extension of basic problem-solving/memory skills; preceiving it to be a chore only dumbs down one's approch toward reasoning information. " I think there are as many ways of organising internal and external information as there are people, and for some people there are more exciting ways to exercise categorising-brain-muscles than by sorting file folders.
I actually believe that the only laziness encouraged by search engines concerns the deletion of useless items. I suspect that in our electronic lives most of us have a a much higher proportion of clutter to ripe good stuff than we would be willing to accept in our material or even emotional lives. I also suspect that many people are disturbed, or at least distracted, by this clutter.
posted by suleikacasilda at 5:32 PM on October 14, 2004
Over the years I have refined my folder system to create a reasonable level of consistency; normally by thinking hard about and differentiating in various "dimensions" further than by "topic"; such as what the information is to be used for, whether its status will change later, whether it overlaps with other topics, and so on. These days my folder system takes much less time to maintain than it used to, since it has become much more refined. I wish nevertheless that someone would write a Powermarks for files and folders.
Smart Dalek, I really didn't get this: "The effort one places in bookmarking or storing files in a directory is an extension of basic problem-solving/memory skills; preceiving it to be a chore only dumbs down one's approch toward reasoning information. " I think there are as many ways of organising internal and external information as there are people, and for some people there are more exciting ways to exercise categorising-brain-muscles than by sorting file folders.
I actually believe that the only laziness encouraged by search engines concerns the deletion of useless items. I suspect that in our electronic lives most of us have a a much higher proportion of clutter to ripe good stuff than we would be willing to accept in our material or even emotional lives. I also suspect that many people are disturbed, or at least distracted, by this clutter.
posted by suleikacasilda at 5:32 PM on October 14, 2004
Really wish this had Firefox support? Then suggest it to them.
posted by tapeguy at 5:42 PM on October 14, 2004
posted by tapeguy at 5:42 PM on October 14, 2004
Google Desktop Search makes AIM crash for me on Windows XP SP2. I had it happen twice mid-conversation, so I just shut GDS down, and will uninstall it soon.
posted by riffola at 5:45 PM on October 14, 2004
posted by riffola at 5:45 PM on October 14, 2004
No firefox support, and we have no idea if it supports network drives or not. Oh, well. Beta, right?
posted by adampsyche at 6:10 PM on October 14, 2004
posted by adampsyche at 6:10 PM on October 14, 2004
What I want is an OS that doesn't have folders, but tags.
I have the feeling that with the categorisation of meta-data in this way, google are gearing up to a middle-ware shell replacement that does something similar to this.
G-mail paradigms extended to a desktop environment - indexable, searchable, able to be tagged with multiple priorities/sections.
Wouldn't that be nice?
I imagine the 1 gig is for the in situ file indexing - that thing would require quite a bit of elbow room.
posted by Mossy at 7:06 PM on October 14, 2004
I have the feeling that with the categorisation of meta-data in this way, google are gearing up to a middle-ware shell replacement that does something similar to this.
G-mail paradigms extended to a desktop environment - indexable, searchable, able to be tagged with multiple priorities/sections.
Wouldn't that be nice?
I imagine the 1 gig is for the in situ file indexing - that thing would require quite a bit of elbow room.
posted by Mossy at 7:06 PM on October 14, 2004
One of my friends told me it also indexes your browser cache - so he was able to keyword search all of his GFs yahoo mail messages.
Caveat User - but I will still probably use it if/when the linux version is released. I have 300 gigs ofporn music important files stuff on my 4 hard drives, so grep, find, etc. just isn't an option for me.
posted by bashos_frog at 7:18 PM on October 14, 2004
Caveat User - but I will still probably use it if/when the linux version is released. I have 300 gigs of
posted by bashos_frog at 7:18 PM on October 14, 2004
Kindall:
You can get the computer to do much of that. Computers can make pretty good guesses about document titles, author names, and the like. And machines are already better than humans at distinguishing spam from good e-mail; there is no reason that computers can't create "topic clusters" of related documents using similar techniques.
None of which deals with the problem I described. The same document might be submitted to 4 different publication venues over the course of its evolution with slight, but non-trivial changes in formatting. "Topic clusters" and making inferences about titles from content will not enable me to tell the early draft version I presented at an informal conference, from the late draft version that was included as a book chapter. However, because I have added metadata to those files (in terms of grouping them into folders labeled by event or publication) it is trivial for me to find which one I really want.
Creating deep hierarchical organization for our private document collections is something geeks do reflexively, and so they think that it is a fundamental component of thinking, because obviously they are smart and anyone who doesn't think like they do isn't thinking right.
Of course, I never said anything about "not thinking right." That is an assumption that you made the leap to on your own. What I said is that full-text searches will not eliminate the need for many users to do some work in providing meta-data.
Besides, this isn't a geek thing. About once every two years I'm tempted to jump ship and start a career as a professional writer. Which has led me to reading a lot of material about how professional writers work. In addition to the creativity side of things, most professional writers argue for developing a rich and structured system for organizing their work in progress. They may use pen and notebooks. They may use typewritten documents in file folders. But most of them say that if you are going to make a business out of writing, you had better be able to track your product through multiple drafts and submissions to its published form. The people who are successful at writing for a living are not just creative, but they good at managing the business of writing as well.
People like shallow hierarchies (i.e. documents within project folders, rarely deeper), and they like to keep what they're working on at the moment in plain view. There is no OS that fully supports the way regular people work.
I'm wondering who you mean by people here? And if you are arguing a strawman here. Nowhere did I say "deep heirarchies," and in fact, the two examples I describe are quite shallow. I use a one folder/project system. While Twyla Tharp uses a big box for each project.
But come on here, while the 7+/-2 rule may be wrong in rigid application, it is correct in principle. Too few items in a list, and there is no point in having a list. Too many items in a list, and stuff gets lost in the noise. That breakpoint may be dependent on what type of list we are talking about, but I know I get frustrated when I have scroll through large directories.
In fact, google seems to get what you don't. Location information is important metadata. As an example, try googling metafilter "don norman" vs. site:metafilter.com "don norman". Or try freebsd dns vs. site:freebsd.org dns. The beta google desktop search has a filename keyword.
A good litmus test to uncover anal-retentive hyper-geeks: they refuse to let iTunes keep their MP3 files organized for them, because they have painstakingly organized their music files in a way that makes sense to them.
A stupid example. The reason why iTunes can find MP3 files so quickly is because someone took the time to create CDDB entries for a large number of the commercial music CDs that are available. Because I tend to favor local music, I find that I frequently need to tag my own files. In addition, there are whole arguments about ID3 tags being insufficient for live performance recordings, jazz and classical music.
Hyper-geeks are big into sunk costs. "I spent all this time creating my own organization for my files, therefore I will further waste my time in increments of one second, because if I don't, I would have wasted the time I spent organizing the files." Given a choice between wasting X units of time, and wasting X+N units time, they will choose to waste X+N units of time "to avoid wasting the X units of time."
Well, here is a practical example. I have an old paper with the title "The Best Things in Life are Free" that I'm thinking about dusting off and using somehow. So, I enter the title as a phrase search into the fulltext search engine that I set up about 6 moths ago. The search comes back with 32 hits.
Which one do I want? Well, I look at the filenames. "Backup of..." (the top rated hit) is probably not the one I want. I don't want any of the 7 pdf versions, (one of which is on my web page, forgot that was there.) It comes up as a hit in my resume. I can ignore that. The fact that I spend 2 seconds of time giving my files meaningful names, and putting them into project directories, means that I don't have to click through 32 different flavors of "Document1.sxw" and "paper.doc."
Of course I could have just double-clicked on projects, and double-clicked on Open_Source_Paper to get there, or I could have simply typed "cd pro[tab]Open[tab] and gotten there even faster than clicking, but instead, I'm supposed to find it "easier" to hunt for the magic combination of search terms that give me the file I want.
With a search engine, any metadata that will cut the number of potential "hits" is worth adding. Any metadata that will make the file you want stand out from a dozen false-positives is worth adding.
posted by KirkJobSluder at 7:23 PM on October 14, 2004
You can get the computer to do much of that. Computers can make pretty good guesses about document titles, author names, and the like. And machines are already better than humans at distinguishing spam from good e-mail; there is no reason that computers can't create "topic clusters" of related documents using similar techniques.
None of which deals with the problem I described. The same document might be submitted to 4 different publication venues over the course of its evolution with slight, but non-trivial changes in formatting. "Topic clusters" and making inferences about titles from content will not enable me to tell the early draft version I presented at an informal conference, from the late draft version that was included as a book chapter. However, because I have added metadata to those files (in terms of grouping them into folders labeled by event or publication) it is trivial for me to find which one I really want.
Creating deep hierarchical organization for our private document collections is something geeks do reflexively, and so they think that it is a fundamental component of thinking, because obviously they are smart and anyone who doesn't think like they do isn't thinking right.
Of course, I never said anything about "not thinking right." That is an assumption that you made the leap to on your own. What I said is that full-text searches will not eliminate the need for many users to do some work in providing meta-data.
Besides, this isn't a geek thing. About once every two years I'm tempted to jump ship and start a career as a professional writer. Which has led me to reading a lot of material about how professional writers work. In addition to the creativity side of things, most professional writers argue for developing a rich and structured system for organizing their work in progress. They may use pen and notebooks. They may use typewritten documents in file folders. But most of them say that if you are going to make a business out of writing, you had better be able to track your product through multiple drafts and submissions to its published form. The people who are successful at writing for a living are not just creative, but they good at managing the business of writing as well.
People like shallow hierarchies (i.e. documents within project folders, rarely deeper), and they like to keep what they're working on at the moment in plain view. There is no OS that fully supports the way regular people work.
I'm wondering who you mean by people here? And if you are arguing a strawman here. Nowhere did I say "deep heirarchies," and in fact, the two examples I describe are quite shallow. I use a one folder/project system. While Twyla Tharp uses a big box for each project.
But come on here, while the 7+/-2 rule may be wrong in rigid application, it is correct in principle. Too few items in a list, and there is no point in having a list. Too many items in a list, and stuff gets lost in the noise. That breakpoint may be dependent on what type of list we are talking about, but I know I get frustrated when I have scroll through large directories.
In fact, google seems to get what you don't. Location information is important metadata. As an example, try googling metafilter "don norman" vs. site:metafilter.com "don norman". Or try freebsd dns vs. site:freebsd.org dns. The beta google desktop search has a filename keyword.
A good litmus test to uncover anal-retentive hyper-geeks: they refuse to let iTunes keep their MP3 files organized for them, because they have painstakingly organized their music files in a way that makes sense to them.
A stupid example. The reason why iTunes can find MP3 files so quickly is because someone took the time to create CDDB entries for a large number of the commercial music CDs that are available. Because I tend to favor local music, I find that I frequently need to tag my own files. In addition, there are whole arguments about ID3 tags being insufficient for live performance recordings, jazz and classical music.
Hyper-geeks are big into sunk costs. "I spent all this time creating my own organization for my files, therefore I will further waste my time in increments of one second, because if I don't, I would have wasted the time I spent organizing the files." Given a choice between wasting X units of time, and wasting X+N units time, they will choose to waste X+N units of time "to avoid wasting the X units of time."
Well, here is a practical example. I have an old paper with the title "The Best Things in Life are Free" that I'm thinking about dusting off and using somehow. So, I enter the title as a phrase search into the fulltext search engine that I set up about 6 moths ago. The search comes back with 32 hits.
Which one do I want? Well, I look at the filenames. "Backup of..." (the top rated hit) is probably not the one I want. I don't want any of the 7 pdf versions, (one of which is on my web page, forgot that was there.) It comes up as a hit in my resume. I can ignore that. The fact that I spend 2 seconds of time giving my files meaningful names, and putting them into project directories, means that I don't have to click through 32 different flavors of "Document1.sxw" and "paper.doc."
Of course I could have just double-clicked on projects, and double-clicked on Open_Source_Paper to get there, or I could have simply typed "cd pro[tab]Open[tab] and gotten there even faster than clicking, but instead, I'm supposed to find it "easier" to hunt for the magic combination of search terms that give me the file I want.
With a search engine, any metadata that will cut the number of potential "hits" is worth adding. Any metadata that will make the file you want stand out from a dozen false-positives is worth adding.
posted by KirkJobSluder at 7:23 PM on October 14, 2004
Mossy, didn't BeOS do that?
One good thing about hierarchies or metadata is that it allows conceptually related things to go together naturally; if I search for one document and have awesome google-fu, I'll get it and possibly miss out on others that would go well with it.
Metadata or some other structuring mechanism will be with us always.
posted by kenko at 7:48 PM on October 14, 2004
One good thing about hierarchies or metadata is that it allows conceptually related things to go together naturally; if I search for one document and have awesome google-fu, I'll get it and possibly miss out on others that would go well with it.
Metadata or some other structuring mechanism will be with us always.
posted by kenko at 7:48 PM on October 14, 2004
suleikacasilda:
I think there are as many ways of organising internal and external information as there are people, and for some people there are more exciting ways to exercise categorising-brain-muscles than by sorting file folders.
You see, that's what I really don't get. People make it sound as if setting up a basic set of file folders is work.
It takes a trivial amount of time to create a new folder with a useful name, like (*) "MAA Conference 2005"
It takes a trivial amount of time to double-click on "MAA Conference 2005", and type three key words into the filename "google fanboy confessions.doc"
It takes a trivial amount of time to perform the extra two mouseclicks to open the file.
I mean, dang, I wish my life were so darn easy that I could bitch about something that takes a grand total of 30 seconds a day.)
(*)Metafilter Addicts Anonymous.
posted by KirkJobSluder at 7:54 PM on October 14, 2004
I think there are as many ways of organising internal and external information as there are people, and for some people there are more exciting ways to exercise categorising-brain-muscles than by sorting file folders.
You see, that's what I really don't get. People make it sound as if setting up a basic set of file folders is work.
It takes a trivial amount of time to create a new folder with a useful name, like (*) "MAA Conference 2005"
It takes a trivial amount of time to double-click on "MAA Conference 2005", and type three key words into the filename "google fanboy confessions.doc"
It takes a trivial amount of time to perform the extra two mouseclicks to open the file.
I mean, dang, I wish my life were so darn easy that I could bitch about something that takes a grand total of 30 seconds a day.)
(*)Metafilter Addicts Anonymous.
posted by KirkJobSluder at 7:54 PM on October 14, 2004
KirkJobSluder, I'm so sorry your stuff is so easily organised into a basic set of file folders. That can't be very interesting for you.
posted by suleikacasilda at 2:49 AM on October 15, 2004
posted by suleikacasilda at 2:49 AM on October 15, 2004
My Mother doesn't have this kind of access to my hard drive.
Why would I give it toThe Department of Homeland Security Google?
posted by fullerine at 4:45 AM on October 15, 2004
Why would I give it to
posted by fullerine at 4:45 AM on October 15, 2004
That can't be very interesting for you.
Sulekacasilda, I'm so sorry your thoughts are so disorganized that you can't make basic distinctions between one general subject and another. (Or, for that matter, create mental associations between subjects without a search engine to assist you.) Life must be very confusing for you.
Think of it this way, guys: You're taking a trivially sophisticated tool (Google Search) and using it to perform a task that was previously performed by the most sophisticated search and association tool that is ever known to have existed (the human mind). Which do you think is liable to find more cretive and interesting connections?
posted by lodurr at 6:18 AM on October 15, 2004
Sulekacasilda, I'm so sorry your thoughts are so disorganized that you can't make basic distinctions between one general subject and another. (Or, for that matter, create mental associations between subjects without a search engine to assist you.) Life must be very confusing for you.
Think of it this way, guys: You're taking a trivially sophisticated tool (Google Search) and using it to perform a task that was previously performed by the most sophisticated search and association tool that is ever known to have existed (the human mind). Which do you think is liable to find more cretive and interesting connections?
posted by lodurr at 6:18 AM on October 15, 2004
A very thought provoking thread - one of the many mefi threads i chose to peruse in the last several years that is now indexed for super-fast searching right here on my desktop.
Installed more than 24hrs ago now, indexed everything by this morning - under the careful scrutiny of Zone Alarm - nothing was sent back to GoogleHQ except the confirmation of a successful install. Exactly as stated in the policy.
I hate MS search it's slow and dumb. This thing is easy to use and so familiar of interface. It's peachy keen. Sure I'll keep a close eye on it for paranoia's sake, but so far I'm a happy clam!
posted by dorcas at 7:42 AM on October 15, 2004
Installed more than 24hrs ago now, indexed everything by this morning - under the careful scrutiny of Zone Alarm - nothing was sent back to GoogleHQ except the confirmation of a successful install. Exactly as stated in the policy.
I hate MS search it's slow and dumb. This thing is easy to use and so familiar of interface. It's peachy keen. Sure I'll keep a close eye on it for paranoia's sake, but so far I'm a happy clam!
posted by dorcas at 7:42 AM on October 15, 2004
I went ahead and created a Firefox-Mycroft search plugin for Google Desktop Search (or is it a Mycroft-Google Desktop Search plugin for Firefox?). It's just a simple little search plugin, but at least it displays 100 results per page by default.
Self-link.
posted by quasistoic at 10:09 AM on October 15, 2004
Self-link.
posted by quasistoic at 10:09 AM on October 15, 2004
« Older Dead or alive - who decides | Internets? Newer »
This thread has been archived and is closed to new comments
I do like the ability to include emails, chats and binary MS Office documents as part of the normal searching. Too bad it's a Windows only application.
posted by rks404 at 7:47 AM on October 14, 2004