I don't know much about this kind of stuff, but what I've picked up is that in France the DMCA doesn't apply. Why should they need to suffer the consequences of other countries' laws?
Image ads have been opt-in since they were introduced. I hope they stay opt-in. For what it's worth, there's also an option to block specific ads although that won't do a lot if all image ads turn into YOU MAY HAVE ALREADY WON seizure beasts.
Clarification: I know that this is about supporting a bill, but it wouldn't make a lot of sense to have different attitudes inside and outside of the company. I admit that I didn't see the note about "approved to be posted publicly by PR" so I was wrong in the "stop posting internal emails discussing these rules publicly" jab.
My major point wasn't arguing that RSS is the shit for parsing data but that it's easy and popular enough to re-use while being widely supported for delivering serial data. Why did I say "serial data"? Because that's what it's being used for. Why did I say that it's useful for notification? Because that's what it's being used for.
Very few things inside of any RSS spec dictate that any RSS feed must be fetched periodically (there are some more or less standard elements to specify when or how often the feed may be fetched, for instance), or that it must be used for any of notification or reading the syndicated articles in particular. RSS feeds themselves are just convenient containers of data, easily parsed. (And yes, I know that parsing invalid RSS feeds of various origins is a science in itself.)
You're completely right that it has no built-in sense of if part x is related to part y, it simply doesn't need to. The need to point these things out are not yet apparent and useful; the loose connections that can be worked out using timestamps and categories/subjects work for now.
The fact that you bring up Usenet and email makes me believe that you've completely missed the point. RSS feed reading/aggregation for me means being able to read stuff from lots of different sources; Usenet and email are both primarily means of discussion. Content notification via email is just a side gig. Usenet and email gives you the ability to participate in the discussion because that's what it was designed to do, RSS doesn't because it wasn't. There's no shortage of exotic usages of RSS feeds out there, like Gmail's Atom feed for incoming mail, but it's not the primary usage, at least not for me, and I would be prepared to wager that it's not the primary usage for the majority of other users either.
As a big company, they'll have to have rules internally on how to deal with these issues. They'll "stay the fuck out of your life" when people stop posting internal emails discussing these rules publicly. I for one think it's very interesting, but apparently it's not an option to you not to read it. Oh wait.
Let's see. If you were given the assignment to "parse a web page for data", what would you do? Hell, let's make that "parse anything for data". The first thing you'd do is that you'd find out some tell-tale signs of where information starts and ends. This could be different on different sets of data, or it could be consistent; on a web page, it'll almost certainly be inconsistent between these pages. So what RSS (and Atom, another similar but more extensive format with the same goal that falls under the same buzzword) is is simply an easy format to deliver serial data in. It's not designed to be "portable" like PDF if that's what you're alluding, and it's certainly not designed to be readable from a text editor. It's designed to be easy to parse while containing as much data as possible about each entry in the feed and the feed itself. This is the technology side.
The application side of it all is that you get notified when your feeds change, because most applications continuously check on these feeds and work out what entries are new or updated since last time. These applications also make it a lot easier to be effective, since the process of checking of serial postings on multiple web pages gets streamlined by reading their feeds instead. This isn't for everyone either, but it's heaven in an executable for those who want to stay on top of things, which includes a lot of people.
But what the article did say about defragging was that it would take place "automatically" "if it is required." That's not the same thing as writing contiguous files in the first place, or some other sensible scheme that makes defragging unnecessary.
It sounds like they'll do the same as OS X's implementation, in which every file below 20MB will be "defragged" - rewritten continiously to disk - when it's opened, read from or written to, if needed. The alternative is of course scheduled defragmentation, which will be a pain, and has been for scheduled indexing and so on in the past. I'm just hoping Longhorn takes the pragmatic route as much as you do.
As for "one file in more than one place", I think this refers to whatever the Longhorn term for Smart Folder is - canned and constantly updated search results. Technically it's not in two places at the same time, it just looks that way in Windows Explorer. To be semantically correct and in tandem with the file system, these folders should contain *shortcuts* to the individual files.
The sad case is that Windows PCs will continue to be dominant for at least the next five years, no matter how this goes. Ignorance of other OSes, apathy towards the whole issue or reluctance to buy computers from a company that "makes the whole widget"** or to support (commies || a widely inconsistent GUI with rough edges and plurality (how many people will get the need to have KDE *and* Gnome?)) will make sure that people stay with Windows.
(** Steve Jobs about Apple at the 2000 MacWorld San Francisco Expo)
I'm as supportive as anyone here of choice and progress, and I think that the alternatives will win ground and that Microsoft are going through a decline, but I also think that the revolution, if any, won't happen overnight, be televised or instantly bring everyone blackjack and hookers.
Google saves history anyway. I'd be very surprised if Yahoo and MSN didn't do this too. They've been doing this for years. IANAL, but it seems to me that Google's privacy policy (http://www.google.com/intl/en/privacy.html) details that they'll only share the info within the company (between different data centers), when required by law and in aggregate (where all personal info is effectively lost). This seems perfectly reasonable unless some dreadful laws pop up all of a sudden - but would Google really be the entity to blame for that?
Letting you view your own history is completely opt-in, you can 'pause' the recording of what appears in your history at any time, and you can clear the history at any time, never mind logging out to hide it. The only privacy issues that can surface here are those that surface because the people using this don't know how to turn it off. How's this any different from the guy surfing porn not knowing how to empty his cache and clear his history? This doesn't seem any more "rife with privacy problems" than your ordinary web browser.
Lock-in is a tough nut to crack; there are open standards that nobody support or use, and there are closed standards that everybody support or use. Lock-in depends *primarily* on popularity and implementation, not on the licensing of your program.
I think there already exists adequate laws and regulations to stop things like abusing a monopoly. (The big shadow on this would be stuff like DRM, software patents and "intellectual property", designed, or at least currently utilized, to work to protect companies and not people, often taking away fair use.) I strongly believe that if you just create good software that does what people want and let these people know, people are going to use your stuff, unless your license or terms stand in the way.
My point isn't that people shouldn't be aware of anti-competitive methods and practices, or licensing; they're both something that everyone developing software and quite a few users should have their heads around at least to a small extent. My point is that if you have great software and you make sure to get the word out, people will use it. I believe that any developer, in any 'camp', if you're a fan of those, should focus on this first and foremost. That's what the users do, after all.
So what you're saying is that "proprietary applications" should not allow extension of functionality in any way at all? Says who? To what end? How does this benefit the user? How is this not some inane construct of your mind, intended to make non-proprietary alternatives look better?
People are free to make their software open source, but people are also equally free to make their software proprietary. I don't see why the extensibility and functionality of your software should hinge on people's perception of your license and/or philosophy and/or business model.
This binary classification scheme of you frankly gives me the willies.
That's true, but maybe they take different stances on some of the decisions Spotlight took. Things like only one result per file might be limiting in some cases. (I have no previous experience of either Spotlight or Google Desktop Search or their respective innards, myself, so I might be talking out of my ass.) There's also bound to be situations where a summary of the "local results" inline with Google Web results will make all the difference. Who knows.
Toolbar point taken - I've never actually tried it myself but I have seen the page. Official stances on the other-platforms issue seem to be an attempt to stifle the emails at best. In hindsight, I can't actually back up that claim on a Linux version as I can't find the place I read that. However...
"Google Desktop Search is a free, downloadable application that allows users to search through e-mail, instant messages, text files and the World Wide Web simultaneously.
Mac users shouldn't hold their breath, however. Google executives told Reuters that while the company plans to develop an OS X-compatible version of its new search tool, it has to be rebuilt from scratch for the Mac operating system." --MacNewsWorld, October 1st 2004
But they do have an API so that you can extend their closed-source application for them. All APIs are about extending or embracing the usage of something; making it appear as any one person or company is having everyone else do the work for them is pessimistic and biased at best - as is overly hyping the potential, of course. But at any rate, it's a good thing that they have an API - there are already OpenOffice.org/StarOffice plugins. No, it's not official, but it's created by someone because they needed it, just like the official, bundled plugins were created by Google employees because *they* needed it.
Google Toolbar: IE/windows only. They actively resist efforts by Mozilla to make a full-featured version including pagerank. They "actively resist" because they already link to a Firefox alternative with much the same functionality. Sounds like an endorsement to me.
Reportedly, Google have announced plans for Mac and Linux versions of Desktop Search, and Desktop Search recently gained support for Trillian logs, Firefox cache (that's even out of the box), even if we're conveniently skipping the plug-in API and repository.
Slashdot Mirror
Per that, is OS X 20 times the OS OS/2 will ever be? ;)
I don't know much about this kind of stuff, but what I've picked up is that in France the DMCA doesn't apply. Why should they need to suffer the consequences of other countries' laws?
Image ads have been opt-in since they were introduced. I hope they stay opt-in. For what it's worth, there's also an option to block specific ads although that won't do a lot if all image ads turn into YOU MAY HAVE ALREADY WON seizure beasts.
My car gets forty rods to the hogshead, and that's the way I likes it!
I'm sure you meant to say "$100K".
"Now even EARLIER than Longhorn!"
Clarification: I know that this is about supporting a bill, but it wouldn't make a lot of sense to have different attitudes inside and outside of the company. I admit that I didn't see the note about "approved to be posted publicly by PR" so I was wrong in the "stop posting internal emails discussing these rules publicly" jab.
no DVD capability at all
From the Mac mini tech specs: "Optical drive Slot-loading Combo Drive (DVD-ROM/CD-RW)".
My major point wasn't arguing that RSS is the shit for parsing data but that it's easy and popular enough to re-use while being widely supported for delivering serial data. Why did I say "serial data"? Because that's what it's being used for. Why did I say that it's useful for notification? Because that's what it's being used for.
Very few things inside of any RSS spec dictate that any RSS feed must be fetched periodically (there are some more or less standard elements to specify when or how often the feed may be fetched, for instance), or that it must be used for any of notification or reading the syndicated articles in particular. RSS feeds themselves are just convenient containers of data, easily parsed. (And yes, I know that parsing invalid RSS feeds of various origins is a science in itself.)
You're completely right that it has no built-in sense of if part x is related to part y, it simply doesn't need to. The need to point these things out are not yet apparent and useful; the loose connections that can be worked out using timestamps and categories/subjects work for now.
The fact that you bring up Usenet and email makes me believe that you've completely missed the point. RSS feed reading/aggregation for me means being able to read stuff from lots of different sources; Usenet and email are both primarily means of discussion. Content notification via email is just a side gig. Usenet and email gives you the ability to participate in the discussion because that's what it was designed to do, RSS doesn't because it wasn't. There's no shortage of exotic usages of RSS feeds out there, like Gmail's Atom feed for incoming mail, but it's not the primary usage, at least not for me, and I would be prepared to wager that it's not the primary usage for the majority of other users either.
As a big company, they'll have to have rules internally on how to deal with these issues. They'll "stay the fuck out of your life" when people stop posting internal emails discussing these rules publicly. I for one think it's very interesting, but apparently it's not an option to you not to read it. Oh wait.
Or "Angry Flowers Considered Harmful".
If nothing else it adheres to Sturgeon's law: "Ninety percent of everything is crud."
Let's see. If you were given the assignment to "parse a web page for data", what would you do? Hell, let's make that "parse anything for data". The first thing you'd do is that you'd find out some tell-tale signs of where information starts and ends. This could be different on different sets of data, or it could be consistent; on a web page, it'll almost certainly be inconsistent between these pages. So what RSS (and Atom, another similar but more extensive format with the same goal that falls under the same buzzword) is is simply an easy format to deliver serial data in. It's not designed to be "portable" like PDF if that's what you're alluding, and it's certainly not designed to be readable from a text editor. It's designed to be easy to parse while containing as much data as possible about each entry in the feed and the feed itself. This is the technology side.
The application side of it all is that you get notified when your feeds change, because most applications continuously check on these feeds and work out what entries are new or updated since last time. These applications also make it a lot easier to be effective, since the process of checking of serial postings on multiple web pages gets streamlined by reading their feeds instead. This isn't for everyone either, but it's heaven in an executable for those who want to stay on top of things, which includes a lot of people.
But what the article did say about defragging was that it would take place "automatically" "if it is required." That's not the same thing as writing contiguous files in the first place, or some other sensible scheme that makes defragging unnecessary.
It sounds like they'll do the same as OS X's implementation, in which every file below 20MB will be "defragged" - rewritten continiously to disk - when it's opened, read from or written to, if needed. The alternative is of course scheduled defragmentation, which will be a pain, and has been for scheduled indexing and so on in the past. I'm just hoping Longhorn takes the pragmatic route as much as you do.
As for "one file in more than one place", I think this refers to whatever the Longhorn term for Smart Folder is - canned and constantly updated search results. Technically it's not in two places at the same time, it just looks that way in Windows Explorer. To be semantically correct and in tandem with the file system, these folders should contain *shortcuts* to the individual files.
The sad case is that Windows PCs will continue to be dominant for at least the next five years, no matter how this goes. Ignorance of other OSes, apathy towards the whole issue or reluctance to buy computers from a company that "makes the whole widget"** or to support (commies || a widely inconsistent GUI with rough edges and plurality (how many people will get the need to have KDE *and* Gnome?)) will make sure that people stay with Windows.
(** Steve Jobs about Apple at the 2000 MacWorld San Francisco Expo)
I'm as supportive as anyone here of choice and progress, and I think that the alternatives will win ground and that Microsoft are going through a decline, but I also think that the revolution, if any, won't happen overnight, be televised or instantly bring everyone blackjack and hookers.
Or the very old one:
We'll fix it in software.
That does absolutely nothing for me.
Google saves history anyway. I'd be very surprised if Yahoo and MSN didn't do this too. They've been doing this for years. IANAL, but it seems to me that Google's privacy policy (http://www.google.com/intl/en/privacy.html) details that they'll only share the info within the company (between different data centers), when required by law and in aggregate (where all personal info is effectively lost). This seems perfectly reasonable unless some dreadful laws pop up all of a sudden - but would Google really be the entity to blame for that?
Letting you view your own history is completely opt-in, you can 'pause' the recording of what appears in your history at any time, and you can clear the history at any time, never mind logging out to hide it. The only privacy issues that can surface here are those that surface because the people using this don't know how to turn it off. How's this any different from the guy surfing porn not knowing how to empty his cache and clear his history? This doesn't seem any more "rife with privacy problems" than your ordinary web browser.
It doesn't stop there. If I cancel my account, will my history be history?
Lock-in is a tough nut to crack; there are open standards that nobody support or use, and there are closed standards that everybody support or use. Lock-in depends *primarily* on popularity and implementation, not on the licensing of your program.
I think there already exists adequate laws and regulations to stop things like abusing a monopoly. (The big shadow on this would be stuff like DRM, software patents and "intellectual property", designed, or at least currently utilized, to work to protect companies and not people, often taking away fair use.) I strongly believe that if you just create good software that does what people want and let these people know, people are going to use your stuff, unless your license or terms stand in the way.
My point isn't that people shouldn't be aware of anti-competitive methods and practices, or licensing; they're both something that everyone developing software and quite a few users should have their heads around at least to a small extent. My point is that if you have great software and you make sure to get the word out, people will use it. I believe that any developer, in any 'camp', if you're a fan of those, should focus on this first and foremost. That's what the users do, after all.
So what you're saying is that "proprietary applications" should not allow extension of functionality in any way at all? Says who? To what end? How does this benefit the user? How is this not some inane construct of your mind, intended to make non-proprietary alternatives look better?
People are free to make their software open source, but people are also equally free to make their software proprietary. I don't see why the extensibility and functionality of your software should hinge on people's perception of your license and/or philosophy and/or business model.
This binary classification scheme of you frankly gives me the willies.
That's true, but maybe they take different stances on some of the decisions Spotlight took. Things like only one result per file might be limiting in some cases. (I have no previous experience of either Spotlight or Google Desktop Search or their respective innards, myself, so I might be talking out of my ass.) There's also bound to be situations where a summary of the "local results" inline with Google Web results will make all the difference. Who knows.
Toolbar point taken - I've never actually tried it myself but I have seen the page. Official stances on the other-platforms issue seem to be an attempt to stifle the emails at best. In hindsight, I can't actually back up that claim on a Linux version as I can't find the place I read that. However...
"Google Desktop Search is a free, downloadable application that allows users to search through e-mail, instant messages, text files and the World Wide Web simultaneously.
Mac users shouldn't hold their breath, however. Google executives told Reuters that while the company plans to develop an OS X-compatible version of its new search tool, it has to be rebuilt from scratch for the Mac operating system." --MacNewsWorld, October 1st 2004
But they do have an API so that you can extend their closed-source application for them. All APIs are about extending or embracing the usage of something; making it appear as any one person or company is having everyone else do the work for them is pessimistic and biased at best - as is overly hyping the potential, of course. But at any rate, it's a good thing that they have an API - there are already OpenOffice.org/StarOffice plugins. No, it's not official, but it's created by someone because they needed it, just like the official, bundled plugins were created by Google employees because *they* needed it.
Google Toolbar: IE/windows only. They actively resist efforts by Mozilla to make a full-featured version including pagerank. They "actively resist" because they already link to a Firefox alternative with much the same functionality. Sounds like an endorsement to me.
Reportedly, Google have announced plans for Mac and Linux versions of Desktop Search, and Desktop Search recently gained support for Trillian logs, Firefox cache (that's even out of the box), even if we're conveniently skipping the plug-in API and repository.
No, but the self-referential RPG Pa-per-View produced by Shigeru Miyamoto will be available at launch time.