It may sound idiotically simple, but according to technology’s leading seer, Apple CEO Steve Jobs, searching for information — not sorting it — is the wave of the future.
Searching for the Perfect OS
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
Being an administrator myself, I tend to know where everything is on my local machine. But I do see this feature speeding up my browsing process! Saving all the clicks to get into the finder and then the proper directory. With just typing the name, Spotlight locating it and then selecting the file opens to proper application and your off.
I think this new feature/technology is work the price of the upgrade by itself! I see a lot of uses for this, and I really can’t wait until first quarter of 2005 to pick Tiger up!
I think it’s great to see that Apple was able to take this concept and actually implement it! Microsoft’s Longhorn, at the latest build I’ve used doesn’t even have it in the OS yet. Saying you’ll do it is completely different from having it and showing a working version of it!
Even though Apple didn’t innovate this concept, but they certainly made it into a functional, easy to use, great technology that will cut browsing to directories in half!
Can’t wait.
I don’t think that this is practical. Sorting is often used to accelerate searches. The part of the Job’s speech that mentions google is very misleading. Google searches so quickly because almost everything is cached. You wouldn’t want to cache your hard drive, even if you could (and most people can’t even anyway).
Microsoft’s Longhorn, at the latest build I’ve used doesn’t even have it in the OS yet.
Its there. It just dosen’t work and turning it off speeds up the OS by a magnitude of 10.
The BeOS file system –which does what Spotlight can do, except searching inside files– was also slow displaying the contents of a big folder. For example, the /boot/home/config/settings/ had about 600 files, and it would take up to 10 seconds to load and display all of them on my dual Celeron 533. This is the price to pay for having metadata/attributes.
However, computers and drives are getting faster and the goodies you are getting from that functionality outweigh the additional slowness you get from such a system (because the file manager has to also read the metadata, not just the filename). Sometimes it can be a little irritating, but the results are good.
It’s not that you can’t find what you’re looking for, but once you’ve got things installed it’s next to impossible to find out what you can get rid of. Most package installers that uninstall either don’t uinstall dependencies or do a half-ass job at it. I’d like a tool that reports back to you, “ok, you or the system has never used these programs or haven’t been used i
oops.
…haven’t been used in x amount of time and here are there dependencies.
Basically, I would like to know about the programs, scripts, and other files that are on the system and how they relate to one another.
Oh, and offtopic, does anybody know of a tool like procviewer, but that will report back the shared libraries that every process is using? There’s a really cool tool for windows called Filemon, but since Linux doesn’t retain backware compatibility the linux version doesn’t work on a modern kernel.
Just another Apple puff piece.
Steve Jobs has introduced this new thing called search to the world! Better alert holders of key domains (google.com, yahoo.com) so they can start thinking about ways to capitalize on this. Also inform the authors of grep and slocate that they can now begin work!
Thank you Steve! Without your vision we would have never have been able to go back in time five years and build a business around search a hundred times larger than Apple’s market cap!!
The Steve Jobs reality distortion field is far ecompassing.
I don’t think that this is practical. Sorting is often used to accelerate searches. The part of the Job’s speech that mentions google is very misleading. Google searches so quickly because almost everything is cached. You wouldn’t want to cache your hard drive, even if you could (and most people can’t even anyway
Google doesn’t cache the entire web, and Tiger won’t cache the HD. In both cases, the system creates an index of the relevant bits of info in a format that it can quickly parse. And it certainly is practical for HDs; you can see it in action in Jobs’ Tiger demo available from the Apple site, as well as the Tiger preview release seeded to developers.
Yay, search wahoo! /sarcasm
The problem is still meta-data. You dont’ ahve enough meta-data to do much searching with. Now search will do fine for text documents, and mp3’s with id3 tags, but what about binaries and zip files and everyhting else. I gues I haven’t seen it in preview or working in front of me, but I doubt this will be as revolutionary with other goodies to actually do a good job with meta-data.
Metadata was the problem with the idea of Gnome’s Storage as well. There isn’t enough to be truly useful. Once that gets solved then search will be easy.
It’s not ‘search’ in general that Steve is talking about, but rather the context and way it’s being used. You can make fun of him if you want, but the way I see it, both Longhorn and Tiger are embracing search and metadata in a way that was only done previously with BeOS (which certainly wasn’t very mainstream).
Also, Linux projects like storage are also embacing local search technology in a similar way. Google too is starting to realize the implications of local search technology, and this is demostrated through gmail. Gmail uses local search to organize and work with mail in new way.
before you can attack those issues you need to build a system that can do it.
Spotlight can do it, all Software developers need to do is make it convenient to add metadata to files you create at the time of creation (save dialogues, importing pics, etc)
if the user can easily apply metadata with out having to go to the file later and add it then the system will work great.
is why are people so against this? should a computer not be able to be a personal secretary for you? should it not be able to give you the exact document or files relating to a project that you ask for?
today we have file name type searches, in the future when Tiger and Longhorn will come out we will have metadata searching and document searching. that means that we are not looking at application centric computing any more, but rather data centric computing. you can set up saved queries and the files you create will be put in those displays automatically. that is what I want, I want the files not the applications.
Why don’t you people actually read the article or see the demo of tiger before you guys talk about something you have not read or seen. It looks great and is very fast. Also if you read the article Jobs did credit Google so I don’t know how you guys can come up with these sarcastic comments and expect them to be taken seriously. Everybody trys to take everything Jobs says or doesn’t say and turn it upside down to try to bash him. He currently runs two very profitable companies that he founded so why don’t you people shutup already with the bashing. He must be doing something right. This from someone who uses a desktop Windows/Linux PC and a Mac laptop so I can see from both worlds.
It will never replace good sorting. Consider a library. If books were anywhere, but the computer could search and tell you exactly where to go, you could get that Stephen King book you know you want. But you can’t go find it and browse at the other books he’s written, or the other authors in that genre. Sure you can type Stephen King and get a list, but that doesn’t let you know that Dean Koontz is right there int he same section and he has similar books.
People like to browse and find things they didn’t know were related. That’s the key. They didn’t realize there was a relation and therefore couldn’t search for it.
Personally, I can’t wait for Reiser to be released.
BeOS had a great idea with BeFS, but as Eugenia said, it was too slow because of the metadate. But with Reiser4, the file itself can be read quickly and easily, while the metadata is stored in a “seperate” file, while the original file itself is the directory containing it.
Now THAT is nice.
XML to organize metadata in a FS? That’s cool, but not when it’s a layer that sits above the actual FS. In Reiser4, if one so wishes, the XML structure is the structure of a directory tree. Though this is already possible, the overhead wouldn’t be that high, and it would allow for some very interesting things.
(Like your mail spool being nice and metadata-ized, or something like gconf just becoming a directory tree for a nice “spatial” traversal)
How can you possibly credit Steve Jobs with being “technology’s leading seer?” If anything, Steve Jobs is very much the opposite. He has even been described (favorably) in business magazines as a CEO whose best skill is to know how to smartly wait and watch. Here it is, 2004, and Steve Jobs is only now willing to proclaim the superiority of search. That is classic Steve Jobs. On a related note, it is also classic Apple-loyalist to credit Steve Jobs with being a seer after Steve Jobs tardily speaks out on an idea whose merits have been understood by the larger community for a long time.
Steve Jobs is not a seer. Instead, he’s a good copier and a good packager. Steve’s bunch at Apple will find some way to make search available as a feature in OS X such that Apple loyalists will believe that OS X’s implementation of search is uniquely cool. That is all. What’s the big deal about that? In other words, where’s the vision?
It boggles my mind how Steve Job’s bunch at Apple can get away with copying their ISV’s products and then be credited with being innovative and have Steve Jobs be credited on this website with being a “seer.” Watson and most recently Konfabulator have been blatantly copied by Apple into OS X. But is anybody on this website crediting the makers of either of those products with being seers or condemning Apple for plagiarism? Not as far as I can tell. Perhaps Steve Jobs’ bunch is able to get away with such plagiarism because Steve Jobs’ fans are so willing to give him a free pass and call him “technology’s leading seer.”
I believe that you need to do better homework, or at least declare a blatant bias before making claims such as Steve Jobs being “technology’s leading seer.” Such claims are not only inaccurate and unfair to the point of being offensive, but really call into the question the credibility of this website. Worse, when you promote the idea that all vision comes from Steve Jobs, then you make it easier for Steve Jobs to plagiarize the efforts of OS X’s ISVs. And in the long run, when those ISVs leave OS X for another community, the platform that you seem so clearly to favor will be hurt.
I have about 600GB of data indexed with my IBM/Lotus Domino Domain Indexer. The server runs under Linux and needs more the 24 hours to index the data on all the drives.
The update of the index does not take that much time and runs scheduled on the Domino server (runing R5 on Red Hat and ND6.5 on Gentoo).
As long as files are saved the way they are saved today, I don’t see a easy solution incoperated into the desktop OS.
Meta-Data is shure a solution, but what about the data on the network? And what about data from diffrend OS?
As long this mess is around, I am happy with my Domino Domain Indexer
Yeah I was thinking some of the same stuff with Reiser4. It could we awesome, and fast!
He still doens’t have the meta-data problem solved in OS X. I watched the video. It seems like one of those nice tech demos, but real world use will be limited I predict. The built in stuff just to integrate sptolight doesn’t seem impressive enough.
P.S. the reason people bash Jobs is that he his a showman, not an engineer. Sometimes he says things that are outright lies. Either that or he doesnt’ keep up with the competition’s offereings. Like the amazing 9.2 mega pixel display , whereas Apple’s best is 4.1 MP. He really does have a reality distortion field, and a hype field. Hype is for n00bs. Not absolving Gates of any of the aforementioned.
I don’t know how Steve Jobs can be a technology seer when he has 0 technical knowledge. He’s never programmed and he’s never done hardware. He had his hand in designing cases. At least Gates was a pretty good assembly language hacker back in the day and was still writing code for Microsoft’s compilers back in the early 80’s.
Coming up with new ideas and knowing what people want and need is what being a seer is all about. Whether that describes Jobs is beyond my knowledge. But there’s no requirement that you must actually sit down and figure out where the screws should go. I seriously doubt that any hacker or hardware monkey will ever be close to being a seer. They just can’t see the forest for all the trees.
OK so the CEO’s of HP, IBM, Dell, Microsoft are all engineers. Yeah whatever. Jobs does over-hype but he helped found the company so I believe he is entitled to that just as Bill Gates is because someone who is there from the beginning will have a closer & more personel relationship with the company then someone who is appointed CEO and had nothing to do with making the company into what it is.
Consider a library. If books were anywhere, but the computer could search and tell you exactly where to go, you could get that Stephen King book you know you want. But you can’t go find it and browse at the other books he’s written, or the other authors in that genre. Sure you can type Stephen King and get a list, but that doesn’t let you know that Dean Koontz is right there int he same section and he has similar books.
That’s what searching metadata is for: searching on the author will give you other books by that author; searching on the genre will give you books of a similar bent; etc.
In a library, that wouldn’t work well because it requires physical motion on your part, and painful searching of call numbers. On a computer, however, it would work quite well.
I will still sort my files religiously, because I don’t trust computer searches (too many bad experiences with the “locate” command). But if they can work out the problems, it’s a great idea.
“Watson and most recently Konfabulator have been blatantly copied by Apple into OS X.”
Not this tripe again. Please, will you please read Dave Hyatt’s Surfin’ Safari blog and honestly say Dashboard is anything like Konfabulator?
http://weblogs.mozillazine.org/hyatt/
Just because widgets tend to look the same and perform similar functions doesn’t mean Apple ripped off Konfabulator.
Clearly, the resemblance between Konfabulator and Dashboard is skin deep. The underlying technology and the PHILOSOPHY of that technology are completely different. Konfabulator requires widget makers to know Javascript wizards (more or less), which is only a little better than the hardcore programming background it took to build the original Apple Desk Accessories. Dashboard allows people to build web pages that look and work like applications. It’s a totally different philosophy.
And so just what would Apple do with Konfab technology if they had bought them? Apple wants Dashboard sit on top of WebCore, just like Safari and Camino and leverage the power of HTML. Exactly what would Apple do with Konfab’s customized Javascript runtime engine? So, let’s be “nice” and buy a company for a technology we don’t plan to use? Oh yeah, that’s a REALLY good business plan.
As for Job being “technology’s leading seer,” it may be more or less hyperbole, but Jobs is at least executing. He forced the labels to be somewhat reasonable with 99 cent music downloads and uniform licensing. Let’s not forget that all the giants of the tech industry were all talk, no walk about digital convergence before iPod and iTunes. Let’s not forget that it was Apple that popularized USB, that popularized WiFi, that made non-linear video editing something almost anyone with basic computer skills can do.
The point is, you don’t have to be 100% original in order to be a “leading seer.” In my mind, to be a leading seer, you have to have the uncanny ability to question why things are done the way they are, and figure out if it can be done better. While this may seem simple to some, it’s actually one of the hardest things to pull off because 99% of the tech industry is focued only on two things: specs and price.
Although he is quite fallible himself, Jobs at least worships at the shrine of user experience and implementation. Just look at how Apple implmented 4-way video conferencing in Tiger’s iChat as opposed to the tunnel-vision thinking of previous multi-person video chat implementations. It’s clear that more people in tech need to question assumptions about how things are done, instead of arguing over who has a bigger hard drive. Definitely not saying Apple is perfect, but the attitude and committment to that vision makes all the difference.
to be lazy… Why do I want to seach for every document I’ve created?! I know what they are and where I’ve put them.
That’s all we need lazy computer users.
I do not recall that Steve Jobs, or anyone else, recommending that everyone save their files randomly. There was no hint of disregarding proper folder/file relationships. What I did hear was that we often get related data placed logically in other folders. A simple search would not necessarily come up with some of these related files. A good meta search engine should be able to find them.
No, the idea is not new. Did Apple steal this idea. No! What they did was capitalize on it by building a reasonably good way of viewing and searching for data without one having to know SQL or other wonderful search languages. Then, it is presented in a rather nice visual fashion. IMOH that is innovating and even MicroSoft is planning to do a similar thing. I do not believe that either company ‘stole’ this idea nor are there any indications that they are copying each other. Out doing or one upmanship, yes. Theft or copying, no.
So, let’s forget the polemics and stick to the topic at hand. Thank you.
Searching won’t replace sorting until computers can intelligently assign meta-data. If I have write a long document with thousands of words, I need the computer to pick out a few key terms and (more important) concepts that will go in the meta-data. If I have to manually assign meta-data then it’s no different from just putting it in a folder/sub-folder /etc… based on those things I chose.
This has already been done on OSX, its called Launchbar.
http://www.obdev.at/products/launchbar/
I guess since Apple ripped them off, its all right to make one for Linux, get cracking boys! I do love this little app, and might actually learn to program just to make one for Haiku or GNU/Linux.
“Sounds like an excuse to be lazy… Why do I want to seach for every document I’ve created?! I know what they are and where I’ve put them.”
And only a small fraction of people are naturally organized enough to use things like Franklin organizers and Palms. Personally, I owned a Palm once and found the thing useless.
Good analogy is how people used to find things on the Internet in 1997 and how they find things now. In 1997, people used portals like Yahoo, with logically organized hierarchies. You went to Yahoo.com and drilled down to the category you were looking for, and then selected one of many websites at the bottom of that directory.
These days, almost everyone uses search, whether it’s Google or Ask Jeeves. The growth in the number of web pages made portals obsolete.
With 120 GB hard drives on personal computers becoming the norm, the same is happening at the user level. It’s not just the files you create, it’s all the files you receive from all over – the stuff your friend’s IM you, the attachments in your emails, the files you download from websites.
The difference in approach is perfectly highlighted by Google’s Gmail vs. Hotmail or Yahoo! Mail. Hotmail and Yahoo thinks it’s about disk space. Google realizes offering 1 GB of disk space is the minimum necessary to make a search-based approach effective. If you have to delete potential items that you might want to search for, you are defeating the very purpose.
So Gmail doesn’t allow even allow you to create folders. It asks you to simply dump everything into a universal Archive and then search for what you need. And guess what? I found my webmail experience improved by an order of magnitude using gmail over a crappy, time-intensive maintenance-required interface like Hotmail.
With Spotlight, Apple is acknowledging that there needs to be the same approach to managing the information on user’s hard drives, without forcing the user to become the janitor.
Google also realizes how important this is, as they are already developing Google search for the desktop.
I, for one, am looking forward to leaving my janitorial duties behind. It was liberating not having to constantly create nested folders for my MP3s when iTunes came along, and I hope the experience will be even better when Spotlight becomes available.
Microsft also realizes that the search metaphor is the thing that will finally replace the desktop metaphor that has dominated OS design for the last 20 years or so. That’s one of the central idea of Longhorn, right?
The question is who can implement the search-centric OS in a way that works in the real world, in a way that doesn’t get in the way of the user. The Spotlight demo at WWDC is a promising sign for Apple.
Paul you make some good points, but IMHO, good housekeeping is good housekeeping. While I’m sure the technology is wonderful, I see this as a technology for those who refuse to learn how to keep their machines in order. Granted, this is probably the majority of users out there, it just seems that this “new” way of thinking just further inclinates people to be sloppy.
Searching for something on the “internet” is far different than looking for something on your own computer. It’s not as if directory structure is that difficult of a concept…
Perhaps Apple should come up with a technology that I can use to search my house… Heaven forbid I should have some order to it and I just keep losing my keys!
In Panther there was already the basis for Latent Semantic Analysis (go Google for it). It does allow a summary of documents to generate metadata as DJ is asking. It’s the basis of Mail.app’s spam filtering, Panther’s “Summarise” service, and Google’s indexing.
Is they are talking about it like it should be an end-all replacement for how a filesystem works. That is a bad call. I know they do not actually MEAN that, but that is how they talk about it. Very misleading.
Of course, the problem with searches is they only really work if you can remember keywords in what you are looking for.
Maybe I’m just getting old but my memory isn’t good enough for that anymore; That’s why I have a computer to keep my thoughts and interests organized and sorted.
Does Spolight mimmic LaunchBar astounding abreviation detection? Like when one launches Photoshop by typing whatever similar to “Pho”, “PTS”, “PS” or whatever or even mistake some keystroke and LauchBar guessing what you wanted to type?
before I forget… hits upon another problem as I see it. All the big producers seem to be so obsessed with finding a way to make the computer easier to use or finding some magical way to revolutionize use… INSTEAD OF SPENDING TIME MAKING HOW MILLIONS OF PEOPLE WORLDWIDE USE IT DAILY WITH NO PROBLEMS WORK BETTER AND MORE RELIABLY!!!
Well, I don’t know if it will inclinate people to be sloppy. Like you I know where my files are and where I have put them. But I rarely know their contents. I’m writing reviews on different subjects and sometimes I don’t remember if I have already written about something. Since my files are named by the issue of the magazine I am writing for, knowing where the files are doesn’t help me. I need to know what’s inside the files.
For the moment I use grep in the terminal. Spotlight will be welcome.
xnetzero, I understand that under ideal circumstances, good housekeeping is good housekeeping. But people are lazy in the real world. And even if one is not normally the lazy type, there are times when you are lazy.
Secondly, I disagree that searching your hard drive is that much different from searching the Internet. Maybe that was true in the Win 3.1 and Mac OS 6 days, when computers were generally not networked. But these days, a typical user receives files and information from a myriad of sources, a huge amount that ends up on your hard drive.
Example. ICQ has its own directory and structor for storing downloaded files, which differs from the default download directory for MSN Messenger. IM is a primary way my friends send me things like photos. Should I have to waste time remembering where each apps stores the file transfers, and furthermore, waste time organizing my file transfers?
Furthermore, what percentage of ordinary users know how to change the default download directory, or even know where to find it? Most of my friends simply click on the link that appears on the IM window automatically after the transfer is completed, but after that, the file might as well be as good as gone, since they couldn’t navigate their way through the hard drive to find it.
Another example. ON my WinXP system at work, every once in a while, I have to retrieve a particular Illustrator file. For me to navigate to that particular file involes double-clicking 7 times, usually after having to click to hide all my open windows so I can navigate to the hard drive to begin the drilling process. It’s a pain in the butt because I don’t even need to edit the file – usually, I need to get to the directory so I can attach it to an email to be sent out.
Yes, I could create a shortcut, but then my desktop would be littered with two or three dozen shortcuts, and shortcuts themselves were invented because navigation had become a pain in the butt. But shortcuts aren’t flexible in that they are hardcoded to a particular destination.
So consider – I know what the name of the file is. I know that it’s an Illustrator file. With Shortcut, I can just hit the search icon in the corner, type in the name, and BOOM(using Jobs’ way of describing an instant effect), the file is there. No minizizing or hiding windows. No multiple clicks. No muss, no fuss, get the file, wham, bam, thank you ma’am and goodbye now that I don’t need it.
With just two actions and several seconds, I have access to a file that is deeply buried in my file system, and it’s something I would be able to do with any dozens of other files I have occassional access to, that I don’t need to have pre-knowledge that I will need it sometime in the future.
Third example. Sometimes I will begin looking for a file but can’t quite remember where it was saved because it’d been a while since I last accessed it. When you have thousands of documents, it happens occasionally. So I waste 10 or 20 minutes checking this directory and that, opening this file and that. It’s a waste of time!
But I almost always remember, yeah, it was a Photoshop document that was named “Template ABC” that I created 2 years ago. With Spotlight, I could find that file in a few seconds.
I understand some people are skeptical of Jobs and don’t trust his extravagant claims. But think about it and you’ll realize even if you are highly organized, a search approach will be entirely superior to the OS experience than the current spacial navigational approach provides. Maybe OS X users will get to experience it first at a mass market level, but like the Wired article note, it’s taken far to long for us to get this far.
Why do I want to seach for every document I’ve created?! I know what they are and where I’ve put them.>>
Bully for you then.
But say you’re like me, somebody with a BA in english, 1 class shy of a history minor, and 1 comprehensive exam shy of an MA in english.
Oh, and I’ve also been writing for a webzine I helped found back in 1998.
With few exceptions I have everything I’ve written on a computer since 1992. This means nearly a thousand documents. Some of them (screw you DOS) I couldn’t give very descriptive file names to.
So, even though though I know exactly where they are, I have many documents with file names that don’t help me much.
Say I want to find that great information about income and literacy rates in Victorian England. Knowing my oeuvre as I do, I’ve narrowed that down to about 30 files in about 6 folders.
Gee, opening up 30 files (some of them 30-40 page graduate level term papers) in various folders and skimming through them to find a quote or citation isn’t a good use of my time.
A powerful down to the word search and metadata feature *is*.
Steve Jobs has proved to be a “technology’s leading seer.”. Although not every innovation comes from Apple, they are almost always the first to recognize true innovation and more important to implement it.
Does anyone need examples?
I am happy we have at least one very good OS on this planet, who will follow?
DJ Jedi Jeff spoke my mind.
this stuff need AI binding or it will not be better than the BeOS one. Only better is the abstraction of the file name and the text it include.
All this mean 2 thing:
1 all those working and talking about this know it lack many part, but they know they can already start to port (copy) what was already done and wait someone make an AI for it later so they can copy it.
2 they are all living in a Gödel imcomplete enclosed system and they fail to see they are in a chiken/egg solution that will lead them nowere.
those 8.3 DOS file names, well, you could rename them now that you have a new system…
C:> rename HISTOR~1.DOC “History of Roman Eating Habits.doc”
or whatevah
I agree though, indexed keyword and metadata seaches will help a lot. Throw everything in ~/All Documents/ and let it search from there
This isn’t quite as helpful though for say developers who may have hundreds of project files, which they need to retain in some for of hierarchy, lest repeated file names start to conflict.
Like others of you, I can’t believe there are people posting here who just don’t get it. “Too Slow”, “Will never work”, “All we need is lazy programmers”…
MS gets it, BeOS got it, Google certainly gets it.
This technology will be brilliant. It’s fast, just check out the keynote (I know, we all don’t have G5’s, but you’ll get the idea)…
Sure, the more metadata a file has, the better, but that’s the case for any filing system really, storing files in a certain directory is adding meta data (indirectly) to files.
Also, for those who want to sort their files (which I still will anyway), this will only help. You’ll be able to search for all files of a certain pattern and drop them into a folder… Boths worlds will not collide.
The comment that got me the most was the lazy programmer one. This guy really doesn’t get it, like many programmers/geeks I know.
OS’s like OSX and XP (and now Linux more and more) aren’t built for programmers/geeks, they are build with people in mind (we programmers/geeks being a subset of that). If an OS can’t be used by most of the population, or they struggle with concepts, then that is the OS’s fault, not theirs (well, maybe 99% of the time anyway ;-).
Finding files is one of these things people do. When you don’t know what you are doing, you will store files all over the place due to lots of reasons, alot of different apps have different default folders to store stuff and so on. The idea that people need to learn basic stills is still valid, what I’m saying is that a lot of times, things aren’t as intuitive as we might think (even on OS X ;-)…
I have to help a lot of people all the time with this one, showing them how to browse folders, putting things in certain spots only to find them forgetting how that was done a week later. Users find themselves sorting music, pictures, movies, files and email, some even web bookmarks etc… That’s a lot of work that computers could help with…
This technology (and WinFS when it hits) etc. will be very nice… I hope we see it in Linux very soon…
Why do I want to seach for every document I’ve created?! I know what they are and where I’ve put them.
We should all be so lucky! In our project at work, we’ve got a document tree that’s hitting the limits of the 80GB harddrive that it’s stored on. My dad and has to keep track of the multiple drafts from a dozen or more writers while preparing the final product. A good, efficient searching mechanism would greatly speed up access to important information in these two cases.
Now, the above scenarios take into account people who understand the filesystem hierarchy. What happens when you have someone like my mom, who doesn’t understand that at all? She can’t find her documents in “My Documents” but happily uses Google to find anything she needs on the internet. As far as she is concerned, Google *is* the internet.
I don’t know if anybody made a comment about laze programmers, but I did say something about lazy users…
Okay– I completely agree that it’ll be a great tool to augment a well organized computer however cluttered it is, but coming from an average users perspective it’s only going to make the computer that much more mystifying (easier to use perhaps, but certainly more complicated). Sure, I can find my files, but I have utterly no concept of where my files are!
I’m not saying that everybody needs to pull their pants up and become experts, but damn if I haven’t worked with enough users who have zero concept of what the hard drive is. Maybe that’s what is frustrating me the most… …I feel like OS vendors and even developers are tying to create stupid users.
I guess in the long run it doesn’t matter, as long as it works it works, right?
You need to understand that just because something isn’t invented “in house” does not mean you’re not allowed to perfect it.
Example:
http://www.teslasociety.com/
Example of his work:
http://antwrp.gsfc.nasa.gov/apod/image/0208/earthlights02_dmsp_big….
For Lumbergh – which has never been a secret …
“At least Gates was a pretty good assembly language hacker back in the day and was still writing code for Microsoft’s compilers back in the early 80’s.”
http://www.folklore.org/StoryView.py?project=Macintosh&story=Donkey…
Mr. Gates does many other thing very well.
http://www.gatesfoundation.org/default.htm
Never work.
It will never replace good sorting. Consider a library. If books were anywhere, but the computer could search and tell you exactly where to go, you could get that Stephen King book you know you want. But you can’t go find it and browse at the other books he’s written, or the other authors in that genre. Sure you can type Stephen King and get a list, but that doesn’t let you know that Dean Koontz is right there int he same section and he has similar books.
Can people please stop writing out of their orifices? (and no, I don’t mean their mouths).
To begin with, it ALREADY works. It works already in the beta (and on the demo during the keynote, if you can’t find the beta seed). Not only it already works as in “it does search”, but it also already works as “IT IS USEFULL”!
So “Never work” is hardly justified.
About your library metaphor now. Nothing prevents Spotlight to work in EXACLTY the same way as you mention. Replace books and e-Books and presto! you can search your PDF or other kind eBooks for works of “the same genre” (which is exactly your example), the same author, the same period of original publication. (You can already do much of that in iTunes: you can search by name, genre, composer, etc, or by ALL TOGETHER. Spotlight is many times more refined, though).
Then you can go *even further* and search INSIDE the works, for references to the sames things. How do you find works related to Prague or otters in a library when it’s not on the title or summary of the book? Hint: you DON’T! With Spotlight you can.
Furthermore, what all these people that talk about “folder organizing” fail to grasp is that it imposes ONLY ONE of arbitraty hierarchies upon your data files. You mention the library example. What if you don’t want to find other books with the same genre, but also MOVIES, MP3 and PICS? Do you really advocate storing your a) movies based on Kings novels, b) mp3 soundtracks of them, (c) ebooks of King d) pictures of Stephen King (say you are an obsessed fan) in the same folder as his books, so you can find them? And if you do advocate such a gross things, what if the next day you want to find all your OST mp3s (including but not limited to those of King’s based movies) how you go about it? Checking in many folders? (in your King folder (for The Shining OST, in PKD folder for Bladerunner and Minority Report OSTs, etc).
Furthermore, with Spotlight you DON’T EVEN have to quit your “sorting” and “organizing”! Nothing prevents you to continue doing so! It’s an ADDITIONAL AID, something DIFFERENT, not a replacement.
since nobody is perfect, then we can’t have perfect OS that made by human. Have to tell god to make one for us.
Perfection is nothingness, therefore the perfect OS is no OS at all. Actually this could be a good argument religious types could use to explain why the universe is imperfect, oops being atheist I shouldn’t have said that…
Um, anyway, I’m all for searching and stuff. Location based organisation is no longer relevant with computers. In fact it is the people who refuse to leave behind old habits (which may have been technically required years ago) who are lazy. Spare me this ‘blame the stupid users’ crap.
And, metadata, ey? Seems to me like the data of a file should not be treated differently to this ‘meta’ data. For example, and image file could have fields like Width, Height, and a raw bitmap or some sort or compressed stream which would normally be part of the ‘data’ (and a field that tells you the format of that stream). The idea is to split each piece up so it can be quickly identified and used in a search. Hey we’ve just re-invented the database!
This technology is for both novice users and power users and it doesn’t matter whether you keep your hard drive organized or not. I keep my hard drive organized but does that mean I enjoy having to click 10 times to get to where something is. No, I don’t. So for me it will be great. And for everyone else who is not organized it will do that and also give them the ability to find what they are looking for if they save their files where ever. So what if you know where everything is, do you honestly enjoy getting to it? Can’t wait!!!
Reading some of these responses have me kind of laughing because something occured to me…
Considering the popularity of things like mouse gestures, mouse wheels and the onslaught of idiocy that is l33t, one would half suspect the ‘great unwashed’ of being deathly afraid of going to the keyboard. One of the reasons I actually like the comments on this site is the posters for the most part at least seem to maintain some level of literacy and a willingness to type complete sentences; Something which is greatly lacking across the rest of the monkey-zone we call the Internet.
After years of trying to make everything accessable from the mouse, and years of claims that the keyboard was an outdated and outmoded device that needs to be replaced with something, anything… The endless jokes about Mac users who are surprised they cannot turn the computer on without the keyboard “Why do I need that? I never use it!”
The latest flavor of ‘hot’ upcoming technologies is next to worthless without one.
I just find that absolutely hilarious.
What Apple seem to be doing is exactly what they are good at:
Taking an advanced technoloy and making it accessible for the general public.
BeOS provided the technology but never really did anything with it. Apple look like they are turning it into something useful – i.e. putting into applications.
And no you don’t need a G5 to use it, Eugenia complained that opening a 660 entry folder took 10 seconds on her dual Celeron 533.
My single Cyrix 150 could search and find all 1500 html file in 15-20 seconds. If the technology is designed properly it’ll be *amazingly* fast on modern systems.
right, but it looks exactly like Konfabulator. Firefox’s AIO Gestures uses different code compared to Opera’s gestures, but yeah, they were copied from Opera.
(not to slur AIO Gestures cuz I’m lovin it)
It’s about time we have a proper search facility for our desktop computers. I have files and folders from back in the late 70s, and over the last 25 years or so I’ve changed organizational schemes a few times. If I want to find some notes I took on my father’s old apartment back in 1989, I really shouldn’t have to spend my time popping folders with a crowbar.
Basically, Spotlight maintains an intelligent concordance. That’s what they call an index of the words in a collection of documents that points from the words to where they appear. Altavista, Google and the other search engines maintain web page concordances that let them execute searches quickly. In general, a concordance is MUCH smaller than the collection it indexes, so the overhead of a concordance can be relatively small.
Law firms use concordances to keep track of big cases, where each side tries to swamp the other with raw paperwork.
If you want to play around with a concordance that includes metadata, try going to Amazon. For some reason, no one on this list has mentioned them. They now have what they call “search inside” which lets you search for words inside the large number of books they have digitized. This used to only work for their Catalogs (Beta) feature, but more recently they got it working for books. The metadata is the usual title, author, date published and that stuff that you can also search by using Advanced Search. Otherwise, they sort of bang the searches together and give you lots of stuff.
You actually can find just about everything that Steven King has written, and you can find books that talk about Steven King.
Sherlock sort of provided a concordance for the Macintosh from way back in System 8, but it was lame. It only indexed the first few thousand bytes of each file, so you can’t search for “a far far better thing”, and they would only give you the first 100 occurences, so you couldn’t scoop up the entire set of relevant files.
It sounds as if Spotlight fixes this.
As for metadata. Some of it is already there. Photographs tend to have EXIF data. Music files have various tags. Still, there isn’t a lot of metadata, but there is a reason for this. There aren’t a lot of programs that let you do anything useful with the metadata.
Microsoft Word version 5 supported a variety of document metadata. This included author, topic, keywords and other cruft. It even tried to force you to fill in this stuff for new files by popping up an obnoxious window that everyone learned out to get rid of.
Once Spotlight makes metadata useful, people will turn this kind of feature back on. You downloaded a PDF of a report or product specs, default metadata is the URL, maybe the PDF has an abstract. Then, you fill in a few words about the report. Since doing so, and doing so with some consistency, will reward you by making the file simpler to find, you’ll do so.
Twenty five years later, you have metadata.
Of course, there is now a niche for new products that can develop metadata by doing a semantic or statistical analysis of the cruft on your disk. A lot of web companies, for example Overture, try to do this with web pages. Why not on your home machine?
There is also now a market for a fast browser that makes it easier for YOU to go through 10,000 files and assign metadata using simple keystrokes or mouse clicks. There is no point in cleaning out that old box of papers in the garage if all you are going to do is stuff them back in the box.
There is a market for alternate displays. I’d love a program that shows me what I’ve been working on as a function of time. I find most recently used files features useless since I’ll use several hundred files in a given day. Knowing which files have been used when and with which applications is further metadata that can really help. Proximity of usage in time is not proof of causation, but it is suggestive.
Of course, these products may get sucked into the OS after a bit, but here’s your chance to make a few bucks.
I guess one day guys, windows users might get to use a real OS. Personal experience shows me that OSX is streets ahead of Windows, and probably always will be. Persoanlly I think what Apple should do is make OSX work on CISC chip, and give the whole world the choice to use it’s brilliant OS. Cause once you try it, you will never go back.
http://arstechnica.com/cpu/4q99/risc-cisc/rvc-1.html
So what CPU do you want OS X to run on? Maybe Apple should try OS X on a 68040…
I guess one day guys, windows users might get to use a real OS. Personal experience shows me that OSX is streets ahead of Windows, and probably always will be. Persoanlly I think what Apple should do is make OSX work on CISC chip, and give the whole world the choice to use it’s brilliant OS. Cause once you try it, you will never go back.
I’d love to see OS X on x86 because honestly if it never happens, then I don’t care how good the OS is, I won’t be using it. Unless Apple starts up the clones again. I’m too sold on commodity hardware at this point.
I reckon I’ll go Linux long before I go OS X at this rate.
“It’s not that you can’t find what you’re looking for, but once you’ve got things installed it’s next to impossible to find out what you can get rid of. Most package installers that uninstall either don’t uinstall dependencies or do a half-ass job at it. I’d like a tool that reports back to you, “ok, you or the system has never used these programs or haven’t been used i”
Windows XP does this with “desktop shortcuts”
Personally I think it is one of the most irritating features I have yet seen. Especially since what it reports back as items I haven’t used, I have in fact used…it is just flat out wrong. I would hate to think of an OS telling me that I haven’t used program XYZ, when I have. That could create real havoc. It might be useful to people who don’t really use their computers, but to power users I would think this would be considered just obnoxious.
“Meta-Data is shure a solution, but what about the data on the network?”
Apparently Apple thought of that.
http://www.eweek.com/article2/0,1759,1618106,00.asp
From the the linked article:
“Spotlight local-file search engine, can be extended over a network, Apple officials told eWEEK.com. Developers will be able to create large virtual directories by combining the indexed logs of other machines. The new capabilities will be driven by Tiger’s expanded list of core services, including Core Data, which will offer a built-in SQL database and metadata store.”
It doesn’t look like they are building network searching directly into Tiger (though who knows about Tiger server), but they have considered that a need are releasing APIs to extend the functionality of their indexing system to index remote machines. Probably a lot of this will be handled via Xgrid.
“to be lazy… Why do I want to seach for every document I’ve created?! I know what they are and where I’ve put them.
That’s all we need lazy computer users.”
Not really. The reality is that some people just have difficulty finding and organizing things on a computer. They mave tightly organized physical files…but as soon as you put them in front of a computer, they make a mess. Additionally, it is possible for users to just accidently put things into places where they don’t belong.
These are cases where good search abilities would be helpful…especially for the poor schmuck who can’t always immediately navigate to their files. This has nothing to do with laziness, but rather it is admission that there are some people who just need help, either occassionally or frequently, to find items on their hard drive.
Computers are meant to be a tool. They should conform to us, not us to them.
LaunchBar isn’t quite the same. Though a useful utility, it doesn’t have nearly as robust search abilities as Spotlight.
Sometimes people just plain out forget where they have located a file. For instance, I see a number of our scientists (I work for a rather large research institute) and they accumulate oodles of files from different experiments they have conducted.
You can look at most of their computers and realize they have been fairly organized, but there are times that folders inside of folders of various research projects start building up…then you have things happen where you forget the project name so you can’t just browse to it, but typing in something like “visual communication” into a serach field when most of your work is based on “auditory perception” would certainly be helpful in picking out those few odd files that just elude you because you have looked at them for 5 years. That is why searching is helpful. Occassionally, you just forget where things are, regardless of how organized you might be.
This website is called OSNews, yet the quality of some of the posts is so ridiculously low that it makes reading through some of the threads really painful. Why would people with little or no understanding of the technology involved in this thread discuss it is beyond me. Before opening your mouth, please find out about what has been unveiled at the WWDC (if you can
There is not much I can say in details because I am not allowed to, but some of the technology presented is VERY exciting. Exciting because it is powerful, elegant and open. It empowers the developer.
Regarding Konfabulator (which is a good idea very poorly implemented technically wise -ie memory usage is huge- and in user interaction -those widgets quickly clutter your desk, although I think they are addressing this issue), Dashboard is infinitely more powerful, better implemented and elegant. I can understand that people tend to side with the underdog, but Dashboard is no Konfabulator. The link above from Daring Fireball gives some explanations but even what is in there just scratches the surface.
Spotlight is extremely powerful and well thought out. Compared to WinFS (judging from the information made available by MS), Spotlight is not at the core of a new file system. This different implementation has important implications. Spotlight is independent of the file system, which means that potentially it could play fair with any file system. Think about this a minute. Both implementations have (as always) advantages and drawbacks, but they are genuine advances that will benefit users on both platforms.
Now I am very surprised that some posters do not understand why search is the next thing. Why do you think Google is valued at more than USD 2 Billion?
Why do you think Yahoo is now using its own search engine? Why is MS ready to take on everybody with a new search engine?
The amount of information that we receive every day is enormous. Properly filing and organising this information is taking longer and longer. And even when it is well organised it is still not easy to find or access. Whether you are “lazy” or not does not make a difference. And in any case why should I organise my information if my computer can do it more efficiently?
Folders are a user centric metaphor, your OS does not care! Let the OS organise the data for me. I want to be able to reorganise that data on the fly depending on what I am currently doing -ie smart folders. Do you see the power of this? I do not care where the info is as long as I can quickly and easily access it, and why should I? I want the power without the chore! I want to concentrate on my work not waste my time organising again and again my hard drive.
Finally to comment on what a poster said about Steve Jobs earlier, technology innovation is only half of the story. We are humans: how we interact with technology is at least as important. I will give you an example which is close to my field. FCP is trouncing Premiere in the professional video market, why? Workflow! FCP workflow is a lot simpler and intuitive. The result: things get done faster even on lesser computers. An other example is Motion. After Effect can do what Motion does, but what takes seconds in Motion takes hours in AE (and it has nothing to do with the processor). So this is as much innovation as the underlying technology (and something that Apple had forgotten before SJ came back).
PS: I use all three major platforms and I hate OS fanfoys .
From the article:
What does it matter where a file is stored, as long as you can find it?
I wonder what administrators of corporate networks would say about that?
It is essential for businesses to manage their secrets. To do this they have to know exactly where the files are.
I’m sure there are solutions to this too, but it’s definitely something that has to be taken into account.
Why do I have this weird feeling that the whole computing world is getting completely hopeless when I’m reading stuff like this?
Couldn’t the big heads at the top, among tons of priorities, choose to devise something else than a system encouraging the most hopeless clutter? A system that will be unbelievably complex to implement, impossible to index correctly and completely, require heavy searches for all or nothing and is in the end almost completely useless for the average mortal?
Is it that hard to classify a bit once in a while? On the other hand, how is the indexing system supposed to figure out how to group related files together unless you tell it?
Wouldn’t an application-wise system be easier to maintain than a centralized system trying to do everything right using a huge metadata system that is sure to become bloated or outdated as time passes, no matter how you implement and manage it?
Won’t the common mortal complain it can’t automatically sort out its landscape pictures from the rest of its collection? Won’t it in the end amount to fill endless amounts of tags instead of classifying manually?
Isn’t running any search much slower than viewing a folder’s content? Won’t it dramatically degrade performance, like the average system isn’t slow enough already? Do we really need another layer of inefficiency, and do we really need to root it that deep into the system?
Is it just me, or have software engineers around the world become that retarded lately? Or do they mean that newest feature to replace The Sims in terms of futility and waste of time?
–Martin
It’s not that you can’t find what you’re looking for, but once you’ve got things installed it’s next to impossible to find out what you can get rid of. Most package installers that uninstall either don’t uinstall dependencies or do a half-ass job at it. I’d like a tool that reports back to you, “ok, you or the system has never used these programs or haven’t been used i
>
>
You don’t want to do this [blindly remove dependencies].
Go ahead and do this if you want. just don’t come whining
to us when you screw your system up.
Doesn’t anyone remember On Technology (Mitch Kapor I think), its had a long forgotten index utility that worked miracles on the ole Mac desktop back when HDs were only 100M. Indexing in the background, an applet to do query. IIRC it was Sys7 that finally broke it and the On crowd went on to do high end SW for another platform. Possibly the single best piece of Apple SW I used besides WriteNow, and both written in asm for great performance.
So its been maybe 15yrs since any OS had something like it and it was only $50.
regards
I really don’t know what to say when I read this!
Spotlight is not just about indexing. Indexing has been present in OSs for a very long time (Windows and MacOS). Spotlight does not prevent you from keeping your files exactly the way you want to either.
Please, please find out about what you are talking about before making blanket statements.
Actually, the desktop idea was first more or less fully implemented by xerox at PARC on the alto iirc.
I think the main problem as it is right now is that not many of us are visionaries – so we “do not see what others see”.
I know that there are some here who do not appreciate somewhat non-quanititative (“sissy”???) aspects like vision but it is a very real talent/ability of which very few excel.
And when I say vissionary do not think that I am referring to some mysterious ability but a very systemic ability but who’s mechanism is so complex that at times – many times in fact – it seems non-quantitative more qualitative.
And thats what my rambling ealier about packaging was about. he guy who said that SJ is basically a pachager is very correct because that is the key function of a vissionary. They gather vast, disparate ammounts of information (trends, fads, history, cycles, economic conditions, etc.) and then they say or do something that is an extremely simplified version of the conclusion.
Because people cannot appreciate (rather “see”) the details or the process that led to the ultimate “manifest” of this “vision” they easily brush aside these statements and the people who make them.
That’s one reason why SJ gets a reputation as being stubborn in the way he runs the company(or used to – he seems to have become soft these days). It’s because people hear what he says and they think it’s driven by ego and the such, not knowing that it might actually be driven by an analysis if actual data/info. and the environment, albeit one that would take a long time to explain and breakdown.
I say all this so easily not because I am Apple zealot and a servant of Maestra RDF but because I have just described myself and a couple people I know. Many times I find myself talking above people’s heads not because I am using jargon or I am speaking in tongues but because my statements, while seeminly normal and everyday conversation, are actually a very summarised version of what I am thinking. This usually leads to an hour or so where I get to elaborate and give analogies etc, etc.
Arister, you are probably wondering where this addresses your complaint. Actually its implied throughout my post but to be more explicit. Our brains work by association and thats what Spotlight brings to OS X. A key difference between visionaries (or packegers coz its more generic i.e. “vision” is only one application of “packaging” abilities) is that packagers are actively participant and aware of those associations and are thus able to leverage them in a very powerful way.
But this is a small number of people. Its a talent just like any other talent, however its inherent complexity and the fact that it exists inside one’s mind makes it more difficult to identify objectively.
In the last couple of month’s I have come across a couple of quotations by Einstein which are “chilling” in a strange yet poetic way. To me they have that effect because, knowing that this guy was not renowned for his poetry, you can “see” (get the sense) that this guy was awed, humbled and accepted the complexity of the universe that was unravelling before him and its in depth mystery and beauty (yes those are the words he uses) led him to make this statments. This is the power of packaging – association if you will.
The funny thing is if you took his name off the quotations they would not be out of place in an English Literature/Poetry as many of them barely mention the universe explicitly, only implicitly. You see why it is so easy for a “packager” to speak above people’s heads.
I shall leave as an exercise to others to find such quotations if they are interested. Funny thing is you do not only have t go to Einstein as many other great minds of the past (and present, although somewhat more rare) have a tendency to when humbled by the world they “see”.
DISCLAIMER: This is to keep all the trolls in check who may attempt to divert constructive discussion: I am not equating SJ to Einstein as is obvious in the part of this post where I say “vision” is only one of the manifest abilities of people that are able to summarise/package vast observed phenomena, natural or man-made
He was talking about DA: desktop accessories.
I do not mean to be insulting but like programmer I have to ask whether you actually read the article because the article and my response are not about the Desktop.
Desktop NOT EQUAL TO Desktop Accesories
Desktop NOT EQUAL TO Dashboard
Desktop NOT EQUAL TO Konfabulator
Desktop NOT EQUAL TO Desklets, etc, etc.
Anyway, the desktop is a “space” as in spatial – dashboard and all the above mentioned are applications.
Also, I am nt as well versed about the details of the XEROX-APPLE thing but if in full you mean as we understand the desktop to be today rather than as the understanding was then, then that would have to be Apple. That’s why Apple was able to sue MS – because today’s understanding of desktop is largely Apple although not originally Apple but Xerox.
I hope I made the distinction clear and I do not mean to come off like I am explaining something to a five-year old (if I did I apologise) its just that many people have a tendency to respond to the forum vs. the article or particular posts vs. all postings.
Just the other day people came down hard on me after a link to my blog was made with an article on how to fix Java. Its fortunate that the two loudest critics were actually nice people who apologised after going and actually reading the article. They were also honest enough to concede that their original comments were based on the forum rather than the article.
I think packaging is not the right word.
What SJ has brought back to Apple (among other things) is to put the user at the center of anything they do. How to simplify and enhance the user experience, the user workflow without numbing the product.
For example if a user action can be done in one step instead of three without restricting user options then this is the way to do it. It might sound obvious but it really is not. This is different from a user interface. You need a deep understanding of a process to simplify it. FCP is a good example for software, Airport Express is a good recent example for hardware: instead of a collection of devices and software/drivers, integrate these functions in a small portable device already aware of its environment (Rendezvous). Nothing revolutionary but a much simpler proposition.
If you’ve ever seen the ‘desktop’ of the PARC machines was basically a much of commands… like opening up /bin or /usr/bin on your desktop and clicking them to do things
Thank for good debate/discuss. When I say package I am referring to the mental process of aquiring all this data/info., arbitrating inconsistencies (or “noise”) and removing redundant associations, and at the end of the day coming up with a concise yet accurate as possible as “package” (summary if you will) that is representative of the whole. Another word for the process would be “genericising” (from generic) but I do not know if that is an actual word ;-).
This is not totally irelevent to what you say though – if you concider all the things you mentioned (user actions, applications, user experience) the final product that Apple ships could be concidered a “generic” version of all the intermediate steps that it took to get there including previous products that may or may not have been released to the public and the actual phases of the development process.
BTW, just remembered a term/phrase that is used to describe “talking over someone’s head” as I described earlier – overloaded statement. The unfortunate thing (for others not for me though) is that SJ uses “overloaded speak” i.e. they are no longer just individual phrases its the whole speech.
That’s why the nickname RDF is fitting and many people can relate to it. In many cases though, the RDF is usually substantiated or shown to be true, albeit at a later date when the statements are no longer as “exciting” and mysterious as when first stated.
You can compare this experience to one of an “apprentice” programmer attending a workshop of “master” programmers. He/Shey think they have stepped into the wrong room coz all they hear is talk of “objects”, “artifacts”, “aspects”, “kinds and classes”, “patterns”, “messages” and all that other “good” abstract stuff. I can just see the young guy:
“Excuse, me is this philosophy? I think I am in the wrong place – I was looking for a programming meet?”
“Yes this is it.”
“Well, what time does it start, I was late and I was expecting it to have started already.”
“It has started.”
“So when to you guys start talking about programming?”
“We are. What king os stuff were you expecting?”
“Well… loops, goto, and something called iteration – I was told to expect it but I do not really know what it is. That might not be my fault though coz my friend was tipsy and tried to suggest that it was related to loops and the word for. I may not know much about programming but I do know basic English and I can say for sure that “for” is not the same as loop.”
[All get cramps as they try to suppress laughter]
James Gosling: ” I think Jonathan (Schwartz, I think) here can help you… (giggle)”
[Outburst into laughter]
Guys, nice chatting but I am already late for work. Hope to catch up tomorrow but cannot guarantee coz its Independence Day – July 4.
Ciao,
“Spotlight local-file search engine, can be extended over a network, Apple officials told eWEEK.com. Developers will be able to create large virtual directories by combining the indexed logs of other machines. The new capabilities will be driven by Tiger’s expanded list of core services, including Core Data, which will offer a built-in SQL database and metadata store.”
It doesn’t look like they are building network searching directly into Tiger (though who knows about Tiger server), but they have considered that a need are releasing APIs to extend the functionality of their indexing system to index remote machines. Probably a lot of this will be handled via Xgrid.
Thanks for the info. I ask my self, how they will handle the security issues and the fact that some data needs translators in order to be indexed (If you dont’t have Excel on your system, then how can the indexing service index the content of the Excel file?)
How could a better, faster search not be useful, particularly if this feature can be used on networked drives. I work at an ad agency. We probably have worked with a hundred clients and on thousands of projects over the years. Hundreds of status and contact reports, thousands of emails, memos and letters, thousands of PDF proofs, thousands of files of written ad copy and much more. And each of these files have been authored, titled and organized on our file servers by a different person in a different way. I need to use existing search features all of the time. What if I need to reference a previous submission to a particular awards show while writing a new submission. Will I find it in a folder named after the award show or the folder named after the project that was submitted? Why should I care? Just give me the document now. Every day I waste time clicking through dense hierarchies of folders – and this despite the fact that we have a generally understood system of organization by client, year and project. If developers are working on faster and better search technology, great.
have you ever opened a word dock or excel dock in a text editor?
the Text itself is ASCII and shows up fine. it is the formating that is binary, so you do not need a translator to read the words in a binary document.
you get use to it.
BTW, all those who plan to upgrade to Tiger, I suggest that you turn on Document properties in Office and start using it, and I suggest that you go back and begin adding Meta information to your files (I am sure that many of you with thousands of files will use Apple script to add the data)
Two days ago I posted a link to an article that shows that the idea for Konfabulator is……guess what….. about 20 years old. And where was it concieved – Apple Computer.
So the original Desk Accessories were small scripted applets that an end user without much programming knowledge could build ? Thats amazing. I was under the impression they were written in ASM.
The author of Konfabulator already stated what he was shooting for when he came up with the idea. In his own words — >
Regarding the conspiracy theory about Konsposé and prior knowledge of Dashboard, here’s the scoop. We knew Dashboard was coming, and we’d been told by *many* people that it was being developed as a “Konfabulator Killer”. We never knew the specifics of it other than it was rumored to be the exact same thing as Konfabulator integrated at the system level. We didn’t know that it was going to be touted as part of Exposé, and we didn’t know that their format was going to be closer to a web page than a structured XML file.
As for Apple having Dashboard technology in Copland or Mac OS 9, they didn’t. My idea for Konfabulator was born from wanting to have a really simple run-time environment for people to develop small specialized applications that could look and behave however they wanted. The key point being that it is up to the user to make the cool Widgets. The user would know what they needed, and the user could then create that. The concept had nothing in the slightest to do with Desk Accessories, and it had absolutely nothing to do with Active Desktop. It was about empowering the user to make their own little apps that did what they wanted.
You can read the rest of his blog about it here.
http://www2.konfabulator.com/journal/
Considering the original Desk Accessories were nothing like a Konfabulator or Dashboard applet I don’t know where anyone gets off thinking Desk Accessories did this first. They were small hardcoded apps, not small apps an end user could create which ran scripted in a hosted runtime.
2 things:
1. Semantics
========
I get what you are saying but do you realise that you (and him) are arguing semantics verses the end result. What this guy has done – while it is a good product – is create a framework that allows a higher level of customization than the original desktop accesories but the idea (and product) is and acts the same.
Thats sounds conflicting but it is not. What I mean is if you take the original quotation from Tribble add “with ability to customize” then you have the definition for Konfabulator. But for obvious reasons DA could not have that ability at the time. If they were so different then you would have do modify (delete some stuff or add more) Tribbble’s statements WAY more than just add a simple phrase.
From an end user point of view it does not matter that they can create their own – UNLESS, and this a big UNLESS – Konfabulator guys REALLY thinks that the biggest market for this is development. I may have the wrong opinion but I do not think that most people will be modding these at home. They can but they won’t.
What is going to happen is that a 3rd party market of already created widgets is going to emerge where people with both the want to program AND the artistic ability wil create the products. This is one “oversight” that both Arlo and team + SJ have deliberatley left out because it suddenly takes away the myth that this is truly something for meagre mortals.
Note, there are millions of things around the world that CAN be built by anybody (and no – I am not talking in theoretical terms, rather eeveryday things that a person with reasonable IQ can learn in a couple of hours e.g bake a cake) but because most people want a quality product vs. a simply functional one, they leave it to people who are specialised in each of the areas.
I am a poweruser of computers and will starting my comp. sci. degree this fall therefore i would be one of those people that should not have a problem modding one of these (I now a little c/c++/java/xml/html/css, etc.) but I will not because it is not something that appeals to me, besides the fact that the sight of my own art manifested would result in depression probably ;-).
In fact, the above is a regurgitation of one of the major points of people that often choose macs over pcs – they do not want to modify them (relative obviously but mostly true) they just want to use them. it is a tool albeit one that you can come to love, but that does not take away the tool aspect.
And such will be with widgets. Arlo and team can tout the added framework as “empowering users” but if they are honest with themselves they will realise the real market for widgets is the consumption market (vs. the creation market) and the people that are really empowered by Dashboard & Konfabulator are developers and artists who will be making and selling these things.
If they do not see this then I am sure they will be out of business real soon coz the “tools” market for widgets (vs. the consumption market for widgets) is very limited (Dashboard aside, the result is still the same). If I were them I would get started with building a business around pre-made widgets.
If they are quick to it they could even establish a web site that would become the front runner in such a market although it should look more mainstream (commercial mainstream that is) than their current blog looking site.
2. Automator
========
I know this does not create visual programs/interfaces (at least not automatically/simply) but this IS the solution that will do what Konfabulator claims it will – empower users. This is like writing Applescripts using Lego blocks or something.
Conclsion
======
Widgets and the framework that supports them is a solution for developers (that will benefit end-users) – Automator is a solution for end-users, both power and non-power, that will benefit end-users.
Perfect example of something that has customization abilities and yet moste people worldwide do not use them.
Same with MS Word and the whole Office.
One could point out that VCR have been said to not be easy BUT the reality I have seen is that the cost/benefit factor is too high. Laugh as much as you will but the average person concider it a luxury (not as in expensive but not so important, bottom in prioity of things in and ouside the house) or auxillary device.
Even a 3 step set up process is a bother. They are just not interested. If it was their TV however, trust me they would take time to learn.
I get what you are saying but do you realise that you (and him) are arguing semantics verses the end result. What this guy has done – while it is a good product – is create a framework that allows a higher level of customization than the original desktop accesories but the idea (and product) is and acts the same.
No I’m arguing about the design of Konfabulator vs. Desk Accessories and Dashboard.
The end product does not act the same as desk accessories. The end product can be changed and modified by an end user. That wasn’t possible with Desk Accessories.
Thats sounds conflicting but it is not. What I mean is if you take the original quotation from Tribble add “with ability to customize” then you have the definition for Konfabulator.
It didn’t happen like that so there is no reason to add what was never said or in the original design spec.
Thats like me saying that if Bill Gates would have added a sentence “with GUI” to his description of DOS you’d have the definition of the MacOS.
But for obvious reasons DA could not have that ability at the time. If they were so different then you would have do modify (delete some stuff or add more) Tribbble’s statements WAY more than just add a simple phrase.
I’d rather leave what was said as it is then try and change it into something its not.
From an end user point of view it does not matter that they can create their own – UNLESS, and this a big UNLESS – Konfabulator guys REALLY thinks that the biggest market for this is development.
The Konfabulator guys said thats what they envisioned with the product, empowering end users to create their own widgets. Thats something Desk Accessories did not do.
I may have the wrong opinion but I do not think that most people will be modding these at home. They can but they won’t.
Just because you don’t think anyone will mod their own dosen’t change the fact that to some extent its obvious that Apple looked at Konfabulator and said ‘neat idea, lets do one of our own’.
The rest of your post I’m not responding to as I’m not out to argue about the marketability of the product.
I’m simply pointing out that anyone who thinks that Konfabulator came from Desk Accesories and that Apple didn’t pull some real ideas from Konfabulator for Dashboard really hasn’t investigated the matter very deeply.
I’m simply pointing out that anyone who thinks that Konfabulator came from Desk Accesories and that Apple didn’t pull some real ideas from Konfabulator for Dashboard really hasn’t investigated the matter very deeply.
Sort of funny, since those who believe that Dashboard is a rip of of Konfabulator haven’t investigated at all.
Sort of funny, since those who believe that Dashboard is a rip of of Konfabulator haven’t investigated at all.
Well then enlighten us. So far everyone running their mouth claiming that Apple didn’t copy Konfabulator at all can’t seem to bring up anything that holds water.
I’ve been waiting to hear the arguments and so far the best I’ve heard is some guy trying to add words that were never said to an apple developer’s statements.
Then guys like you who can’t add jack shit to the discussion either.
goodnight.
have you ever opened a word dock or excel dock in a text editor?
the Text itself is ASCII and shows up fine. it is the formating that is binary, so you do not need a translator to read the words in a binary document.
If everything would be that easy, then we would not have a format problem. But unfortunatly the world is not that simple.
I belive you, that you could live with that *hack* for reading Excel documents. But when your business relay on information found in any of those documents, then I am 100% sure you will think diffrend.
Read this. It’s a good beginning:
http://daringfireball.net/2004/06/dashboard_vs_konfabulator
I am not saying that Spotlight does what I said, for all I know, Spotlight knows how to read the document formats.
my point is that you can create a perl script that can rip out the text from an excel or word file. Since that is possible, and all spotlight does is index the words in the file, the formating is not important. given that, Spotlight does not need to be able to read the formating, and no, that is no some weird “HACK”, it is a fact of the file type and a fact of the needs of spotlight.
great link!!! I think you should post that link in any discussion anywhere on the web regarding Dashboard.
Read this. It’s a good beginning:
First off I’d like to apologize about the tone of my last post. I had a tooth ache and was a bit rude in my comments.
I read the link and it still does not present information leading me to believe that Dashboard was anything more than a copy of Konfabulator.
Dashboard and Konfabulator may not be using wholy unique or new ideas but the two pieces of software are very much alike in both design and implementation.
Sure Dashboard uses WebCore, which makes sense as Dashboard is based on Apple technologies and Konfabulator was built to be portable and not rely on MacOS specifics.
I don’t know anyone arguing that Konfabulator was a completely new thing, as with many technologies it brought together many ideas into an integrated design delivered as one usable package.
I don’t consider it a blatent ripoff of Konfabulator, just as I don’t consider the original Windows GUI to be a blatent ripoff of the MacOS. Like the original Windows GUI was inspired from the MacOS, I feel that Dashboard was very much inspired from Konfabulator.
You can look at both Dashboard and Konfabulator and see the similarities. Its pretty obvious. No way in hell Dashboard was a unique and original idea that spawned at Apple. They saw Konfabulator and made their own spinoff of the technology.
I can appreciate where you are coming from by the tone of your post but the reality is two fold:
1.) While it may not have been said by you here others have the opinion that it is a blatant ripoff – Dashboard that is – here and at almost every post that has been made aroung the web including the Konfabulator site itself. So while you may not be saying this this is the current state of affairs (generally speaking) around the web.
2. However when you say Dashboard and Konfabulator may not be using wholy unique or new ideas but the two pieces of software are very much alike in both design and implementation.
have you really thought about why it is like that.
Once again, in my opinion, I point to the functionality defined by the guy that came up with idea at apple in 1981</>. His definition IZZZ defining Konfabulator – MINUS the ability to customize – that is why they look the same, they are compliant to the definition that was conceived at Apple.
However to an end-user or someone just looking at the apps. noone would guess at that ability. Therefore the ability to customize is an OPTIONAL part of the application – you do not have to customize to use widgets. You do not see a big banner when you launch a widget that tells you YOU ARE ABLE TO CUSTOMIZE
No – because that would be stupid and MORE IMPORTANTLY, Konfabulator the runtime and developer components are “seperate” from a widget that a user has downloaded to use (not to modify).
That’s just MS Office has some developer components like the debugger but they are not marketed as one and the same. In fact now that I have said that this is where a lot of the confusion is coming from.
People keep talking about Konfabulator even when they talking about the widgets which are product thaat runs on the platform Konfabulator. That’s just like using the word Java when talking aabout Eclipse, WebObjects, JBuilder or any other applicaation built in or that runs on Java.
Having seperated these two now I think it is easier to say where the copying is and is not happening.
A.) Widgets
=======
Widgets, what they are they are and their purpose has not been copied by Apple because they are 100% compliant with the definition of the person that first concieved them at Apple.
B.) Implementation
============
The ability to customize, if possible from a drop down menu (or any other widget e.g. dialog box) just like you add toolbars to Office or any other modern gui application – then that is a new feature that Konf… has introduced and should be credited to the guys.
However, if is more of a developer-type feature then the end-user “empowerment” that Konf… claims is jsut the usual everday marketing “hype” – hype in that the feature is there but not as is implied in their marketing. It is an added feature but do not market it like my 50 yr old mom is going to be able to do it – market that to the developer commmunity.
Just like MS does not go around marketing the developer components of Office when they sell it – beacuse end-users are not expected to know how to utilise them and Office, to the non-developer market, is not even marketed as a development platform which it is.
However they do market say for a simple example, the ability to add and remove toolbars as one of the standard custoomization features. If my grandma called and said a toolbar she had used on a different computer is not on the current one, I can tell how to add it in two or three actions.
Is this the type of ability that Konf… has – if it does then it is a new product. A last example. In .Net MS is working on how to make it easier to manipulate Office (the ability thaat is previously known as VBA). One of the ways that you caan do this is you can now (or will be in a year or so) actually open Office programs inside Visual Studio and maanipulate them just you would when you build a gui app.
Now tell me, when MS sells Office should they include all of the abilities of VS in their marketing after alll, you can customize any paart of Office using it. I leave that to you guy to respond to but pleaase calaarify when you are talking aabout widgets the applicaation and Konfabulator the platform. Thanks.
Hey SteveB:
“Thanks for the info. I ask my self, how they will handle the security issues and the fact that some data needs translators in order to be indexed (If you dont’t have Excel on your system, then how can the indexing service index the content of the Excel file?)”
Umm…not exactly sure what you are talking about with regard to translators. As far as security constraints across searching and index files of remote systems, Apple has plenty of capability built into their permissions system. You can prohibit access to select files or entire directories if you wish. As a matter of fact, remote access is disabled by default.
They also have access control list support in Tiger. If you are an admin of a large network, this is one of the tools you would likely choose to implement to prohibit or grant access to individuals as needed.