Remember the Battle of Seattle? Who's watch was that on? And that wasn't the start of it.
GWB certainly made use of the FSZ policy and did not reverse the trend, but it's more that many on the left only *noticed* the accelerating trend toward limiting public protest after they no longer had the glare off Bill Clinton's saxophone to blind them.
Here's hoping Obama doesn't get a free pass to do the same stuff just by flashing his grin and speaking in complete sentences. Sadly, I'm not very hopeful.
Exactly. I think part of the problem is that there's not only a social disconnect between 'big IT' and 'small IT', there's also a technical disconnect.
Technically, I think what we really need is something of the order of an 'enterprise applet service', where users can create and publish zillions of teeny tiny spreadsheet-like applications, and have them centrally stored and filed and backed up and expressed in a *standardised*, logically consistent language (which is what Excel and Access do NOT provide, changing incompatibly every version) - but leave the actual 'programming' to the users. That would give the reliability of centralised IT with the flexibility of the Excel zoo.
Socially, we really really need to eliminate this artificial and overbearing class distinction between 'users' and 'programmers'. Because in many cases, it actually is the users who do know best what kind of data manipulation they need to do in order to do their jobs. Surprise! Not really. Turns out smart people actually are smart, but just not about everything all at once.
Read Christopher Alexander, Margaret Wheatley and Dee Hock. There's a lot of low-hanging fruit in that space of 'chaotic self-organising network structures' - but to support it, we need IT infrastructures and methodologies that provide both democracy and interoperability. Excel gives us the first but not the second. Big ERP systems and processes give us the second but not the first. There's a killer app waiting to be born here.
"It has to do with making and enforcing rules like: if too many sheep die, we shoot and skin the wolves responsible. "
Or we could ask, "why exactly do we think that we need wolves to lead us at all?"
I mean, other than that's the way it's always been done, that sheep and wolves have this wonderful symbiotic relationship, that we've got respected sheep-universities turning out degrees in applied wolfhood, that we actively promote at all levels that sheep must aspire to become wolves and if they don't they've got no ambition.
("Do you want to spend your life chewing grass like your mother and I did? Then get out there and sharpen those teeth! I don't want to hear none of this vegetarian talk! Get to like the taste of mutton! Bite! Bite! Bite! And howl! Yeah, that's it! You're doing it! Pursue that Happyness! Go sit the Baa Exam! We're with you, tiger! Hoo-ah! Wait... why are you looking at us with that hungry look? Ungrateful little...!")
"However, I disagree that the markets would not be functioning optimally, while crashing after a long positive run. I would argue that such a crush would serve as a deterrent against repeating the mistakes of the past."
That seems much like the guiding philosophy behind Windows.
"We don't need any security checks at all, because if any user code causes a crash, well, that will just serve as a deterrent against writing buggy or insecure code, that programmer's system will die, and so the quality of software will continually improve.
"Eventually, emergent forces will give us a perfect operating system. And we don't have to do anything to bring it about. We've solved the stability problem! A most elegant solution. QED."
I personally use FF3 with Flashblock rather than Adblock, mainly because I've had some bad experiences with over-zealous filtering in the past. Also, it's really the moving Flash ads that bug me; static ads aren't nearly as annoying.
Except for mouseover hover ads; with those, I just immediately boycott the entire site and never come back. It's really not worth my time and sanity to *ever* see one of those.
(Webmasters, please take note. Want to lose your audience? Hover ads, every time.)
"All of these hair-brained (actually, idiotic) schemes like Carbon Capture are bound to fall foul of the law of unintended consequences."
I tend to agree, though you'd probably class me as a 'warmist'. Carbon sequestration particularly bugs me: a hugely inefficient, inelegant mad-science fever-dream that doesn't even look good on *paper*.
"whatever "it" actually is (assuming "it" exists at all)."
Ice shelves are melting. The Northwest Passage is open. That's not disputable. *Something* is driving rapid warming right now, and that means rapid climate change, and rapid change means trauma.
The question now is do we want to just shrug and say 'it's not proven to be human-caused' and suck it up and suffer guaranteed massive trauma to our civilisation, or do we want to consider modifying our behaviour on the chance it might avert the worst of a catastrophe.
No chance if we don't change versus some chance if we do. Seems like good gambling logic to me.
Indeed? More funding than, for example, the heavy industries which are generating lots of CO2?
There's a lot of political *talk* about fighting global warming, and a lot of righteous head-nodding which could easily be mistaken for genuine will to act. Column-inches are cheap. Rebadging ad campaigns to add stylised flowers and 'eco-friendly' logos, also cheap.
I'll believe there's a huge 'global warming alarmism industry' when it actually *does* something that the *real* power brokers don't like.
"IE updates are managed thru a single interface, windows update, and windows update is actually one small thing windows gets mostly right. I don't want every god awful program under the sun phoning home ON ITS OWN to god knows where and updating itself without my knowledge."
Yes. Firefox is actually a lot harder to administer in a corporate environment because it does its own updates.
I also wish Windows Update could be generalised to trusted third-parties - like Linux updates - so things like Google and Adobe and Apple don't keep trying to manage their own patches.
One thing that does annoy me with Windows Update though, is the 'revision' misfeature. Microsoft keeps 'revising' already shipped patches, which can alter either their code or their detection parameters. This is a very dangerous thing and should not be allowed, because all patches have to go through a testing and approval process here. If a previously approved patch gets revised in such a way that it then automatically installs on systems it was not approved for, it can do a lot of damage.
Linux gets it right: change a package, it's a new version, end of argument.
"The Linux desktop arrived in 1998 with RedHat 6.0."
Yes! Same here.
Actually it was Red Hat 5.0 which I first installed in 1997, but I managed to stuff the boot process and couldn't figure out how to install it until 1998.
It was 1998 when I started dual-booting and then 1999 I think when I went Linux cold-turkey. And it probably was RH 6 I used then.
What makes Windows Explorer, from Win95 to XP, great is that it is customisable, on a per-user and/or per-directory basis. It has a number of orthogonal options which can be switched, and it then remembers that view. Such as: toolbars, show/hide status bar, major mode (thumbnails, list, detail etc), arrange-by, side pane (which remembers its size and is dismissable with one quick X), 'up', 'back'.
These options are great. You don't need to mess with them when your view looks ok, but since you 'live' in a file browser so much, you need to be able to tweak it when things grate.
Nautilus doesn't do this. Instead of options, it has modes. You can't pick and choose the view options *you* want for each folder, you can only pick from a tiny subset of ones the *developers* thought you *should* want.
It's little things like not being to turn off the status bar unless you're in spatial mode, not being able to adjust the size of the side pane, not being able to dismiss the side pane without hitting the menu (because usually you bring up the pane to navigate, then once you're there you need more screen real estate in a hurry) - these little, pointless restrictions just chafe.
There's no reason why the user experience needs to be restricted like this. The 'spatial' argument was where it showed the most. 'The user is using it wrong'. No. That's never the right answer.
It's a design philosophy which needs to be changed.
"Linux could clean Windows' clock if the GUI were more dependable. Right now it's pretty good, but occasionally falls flat on its face."
And please, please, PLEASE can we get (on Gnome) a DECENT replacement for Nautilus.
Nautilus just makes me wince. It's like it's taken half a squint at Windows Explorer circa 1998, on a very bad day, upside down, in a mist, and then looked at Mac Finder and did its best to forget all the good things in Explorer.
Whoever did the original Win95 Explorer design needs a medal: it's still the best feature of Windows. It's simple, intuitive, and it doesn't penalise advanced users. This is what Nautilus should be, but isn't even trying.
(Vista has also done its best to go the same route, and throw out all the things that made Win95's Explorer work.)
These are the silly things that Nautilus does:
1. Half-implemented 'web view' pane. It's useless. If I want to view something as a web page, I'll use Firefox or the browser of my choice (please not Epiphany).
2. No decent tree view. It's tacked on in the side pane, but feels ugly, horrible, restricted. It's not as easy or flexible to use as Explorer's. It's an afterthought.
3. No simple TEXT BOX view of the current location - or if it's available, it's hidden. OSX and Vista have both abandoned this but that's no reason for Linux to. The location needs to be a text box so you can Copy/Paste. That's important for advanced users, because locations are not opaque things that you can 'discover' through a conversation process, but are things you need to *communicate* to other programs and to humans. Text is the only reliable way of communicating, icons don't cut it (you can't cut and paste an icon into an email or IM or config file).
4. 'Emblems'. Sort of cute idea, but implementing anything like this at the gui file-browser level is the Wrong Place to do it. Again, because you can't communicate the presence of emblems - it's metadata that only exists in an interactive browsing session. So you can't share emblems, you can't copy/paste them, then don't exist for anyone but you and only when you're using Nautilus. So useless.
5. No decent 'detail view'. Zoomable thumbnails are sort of okay (though it's very slow to process thumbnails when you're copying a bunch of photographs), but sometimes you really do need to do some serious forensics on a directory and instead of having to drop into command-line, it would be nice to have a somewhat pleasant GUI view of the real files that are there without trying to talk down to you. Nautilus keeps trying to belittle the user and hide them from information 'for their own good'. It's a bad Apple habit, and Windows (pre-Vista) learned not to do it. Stop it.
6. 'Spatial mode'. Nuff said. No, it wasn't innovative, nor was it pleasant. Win95's Explorer had this - as one of two modes that you could select, and advanced users quickly found 'open in same window' much more usable.
Thank goodness Ubuntu hacked it off and made Nautilus nearly usable, but the Gnome folks' response still leaves a nasty taste in my mouth.
"Ultimately there's not $20, there's the $10 I used to have but instead gave to you, and at the end of the day I can at best expect to end up with that $10 back plus interest."
Something I've always wondered: how, exactly, are you going to get that $10+interest back unless at least enough new money to cover the interest is actually *created* rather than just *loaned out*?
Isn't the whole concept of interest-bearing debt really a Ponzi scheme? Eventually it's going to fall over because you can't keep loaning money that doesn't exist and then asking for more back than exists.
Why don't we just forgive all the debts and start again?
"I too disagree with Steven's argument. But people who jump on "tubes" often do not even know the concepts behind the analogy. In a lot of cases, the people that laugh at his comment are even less informed about the topic than Stevens."
Yes! THANK YOU!
I get really annoyed with Slashdot people who laugh at 'tubes' yet still use the word 'pipe'.
More importantly, the Internet infrastructure *does* have non-infinite capacity, and filesharing of movie-sized files, legal or not, *does* cause non-zero consumption of that capacity.
Internet access *doesn't* come for free or even cheap, and a flat monthly access fee *doesn't* necessarily scale to cover all conceivable maximum-bandwidth uses.
In New Zealand and Australia, we pay for our monthly data transfer usage. It's fair, it's simple, and everyone's happy. I don't understand why Americans can't cope with the concept of 'non-infinite supply of a commodity' and 'paying a fair price for what you consume'.
Seriously, the rest of the world looks at you guys sometimes and goes 'huh?'
"Everybody and his uncle tries to make systems that will index every piece of crap on your PC and it invariably results in a useless and horrible waste of resources."
On the contrary, we should seriously be asking ourselves *why*, when all our data is sitting there on our PCs, we've let ourselves get into such a state of disorganisation at the operating system level that a class of program called 'indexer' exists as a third-party tool in the first place.
How come it's not already taken as given that the primary thing an operating system *does* is, you know, *know where all its data is*?
It's as if we're living in an age before 'directories' were invented - or before databases had 'indexes' and 'queries' - and we have to manually write down and key in raw sector numbers every time we open a file. And we're okay with that, because we think - and teach - that that's 'just how computers work'. We've accepted that there's a whole class of things our computers can't do 'because there's no application to do that'.
"I think that is where something like Nepomuk could succeed where internet-wide semantic standards fail to gain traction. People are lazy but they DO devote more effort to organising their own personal data vs. what is on a web app. I do make more effort to tag photos with metadata in F-spot for example. If there is a "structural-level" standard that could be applied to all files so I can tag spreadsheets, photos, databases, addressbook contacts...whatever..and I could follow one simple, consistent process to apply my own metadata and search and organise on that data then it is very useful."
Yes, exactly. I have been wanting something like this for several years now.
I'm sick of how our desktop applications still organise data into little silos that don't communicate. All this data - photos, MP3s, email, IMs, Web bookmarks, text documents - has a fairly rigid structure, or well-defined metadata. So representing it as RDF presents no major semantic problems - the schema's already been defined.
The main problem is infrastructural: we still have this belief that we need to 'run an application' to access a piece of data. I have to 'launch my mailer' to read an email; I have to launch my browser to read a web page; I have to launch Word to read a Word document, and so on. Even if I don't want the whole GUI experience and just want to correlate some metadata. I don't even have a way to *talk* about correlating metadata between Word and my MP3 collection.
We've somehow linked *knowledge* to *human-initiated, application-centred tasks* -- and I blame object oriented programming for spreading this idea by binding 'code' and 'data' together -- but the two are nowhere near the same, and this linkage really limits what I can delegate to my computer.
An MP3 *song* is not an instance of an MP3 *player*, a Word *document* is not an instance of the Word GUI. Nouns and verbs should *not* be tightly bound together. I should be able to analyse the files (nouns) in my system completely independently of the 'user interfaces' (verbs) I can use to manipulate them. But so far, I can't.
Some kind of desktop-wide data framework that can let me construct queries between the structured *data* represented by my emails, my MP3s, my calendar and my photos, for instance, would let me create whole new forms of 'application' or 'applet' on the fly. And blur the distinction between 'user' and 'application developer' in the same way that creating SQL queries blurs the distinction between 'database user' and 'database administrator'.
I've used database systems where users can't construct ad-hoc queries - they're painful, just like today's Windows desktops.
We never should have separated those roles in the first place, and if RDF can take us back to letting *me* decide how to use and interpret *my data*, that's a step forward.
"Regardless of cause, given the failure of semantic markup in such a limited and controlled scope, it is very unlikely that it will succeed with the richness and complexity of all the data in the world."
What the heck is *with* this argument? It's completely false.
Look, we already have exactly the same thing as what the Semantic Web is trying to do: it's called SQL. Oh noes! Trying to describe real world information on a computer will fail! It's impossible!
And yet, impossible or not, we have databases. And your desktop is full of semantic markup.
Your digital camera, that takes photos, did you know it does 'semantic markup'? But it does. EXIF data attaches time, date, shutter settings to your JPG. Your MP3 collection has 'semantic markup' - the ID3 tags show title, author, album. Your email has 'semantic markup' in its headers and in the folder that you sort it into. Your calendar has 'semantic markup' in its time and date fields; if you have iCalendar, you can publish this to the Web.
Wikipedia is full of semantic markup, in its 'category' tags and links and in all those little infoboxes on each page. There's now a Semantic MediaWiki fork which uses RDF to store that data so you can query it.
The Semantic Web is nothing but an attempt to *standardise data representation* so you can do SQL-type queries between data sources you trust but might not reside on your machine: your photo collection, your MP3s, your friends' calendars. It's nothing more magical than web-distributed RDF, which is nothing but SQL restricted to triples, ie, tables of two columns. And we already understand SQL.
Quit the philosophical strawmen and realise that what you claim is impossible is *already happening*.
The reason we don't yet have *massive* RDF adoption is that it takes a while for people to catch on to the kind of crosslinking that can be done between applications; we've been trained by Windows to believe that linking data and creating mashups is *hard* and should only be done by experts. But things like Nepomuk are showing slowly how it can be done.
"The media was already questioning the war long before the WMD claims were debunked. "
Yes, because it was obvious to anyone paying attention long before the war that the WMD claims were outrageous lies. The US intelligence outfits were leaking like a sieve trying to get people to notice the truth. If you followed a war-centric news aggregator like, for instance, antiwar.com (run by Libertarian types, not leftists btw) in late 2002 or early 2003, you could see literally dozens of reports coming out from various highly-placed sources frantically trying to defuse the war before it exploded. Comparing the quality of the information side by side, there was no comparison.
Usually it was small media outlets, local newspapers and the like, which would pick up a quality anti-war story from a top source. Then it would die quietly before it reached the 'big boys', the New York Timeses or the Washington Posts. The amount of coverage the big boys gave to a story seemed inversely proportional to its importance; weird tiny semantic things like the Plame case exploded into big scandals because they weren't directly criticising the *war*, just specific tactics for fighting it.
Watching the whole thing play out actually gave me *more* faith in American democracy. It revealed that there wasn't a big conspiracy controlling everything, because they sure couldn't stop the truth from getting out. There was only a loose network of small conspiracies. All they could do was delay the big media outfits from questioning the 'facts' until it was too late.
Slashdot Mirror
It's official: Tsetse flies are the new obligatory metaphor.
I'm no fan of GWB, but give him credit: the Free Speech Zone policy did not originate with Bush.
Remember the Battle of Seattle? Who's watch was that on? And that wasn't the start of it.
GWB certainly made use of the FSZ policy and did not reverse the trend, but it's more that many on the left only *noticed* the accelerating trend toward limiting public protest after they no longer had the glare off Bill Clinton's saxophone to blind them.
Here's hoping Obama doesn't get a free pass to do the same stuff just by flashing his grin and speaking in complete sentences. Sadly, I'm not very hopeful.
Exactly. I think part of the problem is that there's not only a social disconnect between 'big IT' and 'small IT', there's also a technical disconnect.
Technically, I think what we really need is something of the order of an 'enterprise applet service', where users can create and publish zillions of teeny tiny spreadsheet-like applications, and have them centrally stored and filed and backed up and expressed in a *standardised*, logically consistent language (which is what Excel and Access do NOT provide, changing incompatibly every version) - but leave the actual 'programming' to the users. That would give the reliability of centralised IT with the flexibility of the Excel zoo.
Socially, we really really need to eliminate this artificial and overbearing class distinction between 'users' and 'programmers'. Because in many cases, it actually is the users who do know best what kind of data manipulation they need to do in order to do their jobs. Surprise! Not really. Turns out smart people actually are smart, but just not about everything all at once.
Read Christopher Alexander, Margaret Wheatley and Dee Hock. There's a lot of low-hanging fruit in that space of 'chaotic self-organising network structures' - but to support it, we need IT infrastructures and methodologies that provide both democracy and interoperability. Excel gives us the first but not the second. Big ERP systems and processes give us the second but not the first. There's a killer app waiting to be born here.
"We are going to giggle ourselves to death with LOLcats, and people will argue vehemently that it's morally better than any alternative."
I can think of worse ways to die.
"We're in the death throws of the internet-of-the-corporate-hack."
Are death throws something like Hail Mary passes?
"It has to do with making and enforcing rules like: if too many sheep die, we shoot and skin the wolves responsible. "
Or we could ask, "why exactly do we think that we need wolves to lead us at all?"
I mean, other than that's the way it's always been done, that sheep and wolves have this wonderful symbiotic relationship, that we've got respected sheep-universities turning out degrees in applied wolfhood, that we actively promote at all levels that sheep must aspire to become wolves and if they don't they've got no ambition.
("Do you want to spend your life chewing grass like your mother and I did? Then get out there and sharpen those teeth! I don't want to hear none of this vegetarian talk! Get to like the taste of mutton! Bite! Bite! Bite! And howl! Yeah, that's it! You're doing it! Pursue that Happyness! Go sit the Baa Exam! We're with you, tiger! Hoo-ah! Wait... why are you looking at us with that hungry look? Ungrateful little...!")
"However, I disagree that the markets would not be functioning optimally, while crashing after a long positive run. I would argue that such a crush would serve as a deterrent against repeating the mistakes of the past."
That seems much like the guiding philosophy behind Windows.
"We don't need any security checks at all, because if any user code causes a crash, well, that will just serve as a deterrent against writing buggy or insecure code, that programmer's system will die, and so the quality of software will continually improve.
"Eventually, emergent forces will give us a perfect operating system. And we don't have to do anything to bring it about. We've solved the stability problem! A most elegant solution. QED."
And that's how we got the Internet we have today.
I personally use FF3 with Flashblock rather than Adblock, mainly because I've had some bad experiences with over-zealous filtering in the past. Also, it's really the moving Flash ads that bug me; static ads aren't nearly as annoying.
Except for mouseover hover ads; with those, I just immediately boycott the entire site and never come back. It's really not worth my time and sanity to *ever* see one of those.
(Webmasters, please take note. Want to lose your audience? Hover ads, every time.)
"All of these hair-brained (actually, idiotic) schemes like Carbon Capture are bound to fall foul of the law of unintended consequences."
I tend to agree, though you'd probably class me as a 'warmist'. Carbon sequestration particularly bugs me: a hugely inefficient, inelegant mad-science fever-dream that doesn't even look good on *paper*.
"whatever "it" actually is (assuming "it" exists at all)."
Ice shelves are melting. The Northwest Passage is open. That's not disputable. *Something* is driving rapid warming right now, and that means rapid climate change, and rapid change means trauma.
The question now is do we want to just shrug and say 'it's not proven to be human-caused' and suck it up and suffer guaranteed massive trauma to our civilisation, or do we want to consider modifying our behaviour on the chance it might avert the worst of a catastrophe.
No chance if we don't change versus some chance if we do. Seems like good gambling logic to me.
"and a gravytrain of funding. Don't forget that."
Indeed? More funding than, for example, the heavy industries which are generating lots of CO2?
There's a lot of political *talk* about fighting global warming, and a lot of righteous head-nodding which could easily be mistaken for genuine will to act. Column-inches are cheap. Rebadging ad campaigns to add stylised flowers and 'eco-friendly' logos, also cheap.
I'll believe there's a huge 'global warming alarmism industry' when it actually *does* something that the *real* power brokers don't like.
Not only is water a chemical, it's full of quarks held together with nuclear force.
The stuff's practically a bomb waiting to blow.
"IE updates are managed thru a single interface, windows update, and windows update is actually one small thing windows gets mostly right. I don't want every god awful program under the sun phoning home ON ITS OWN to god knows where and updating itself without my knowledge."
Yes. Firefox is actually a lot harder to administer in a corporate environment because it does its own updates.
I also wish Windows Update could be generalised to trusted third-parties - like Linux updates - so things like Google and Adobe and Apple don't keep trying to manage their own patches.
One thing that does annoy me with Windows Update though, is the 'revision' misfeature. Microsoft keeps 'revising' already shipped patches, which can alter either their code or their detection parameters. This is a very dangerous thing and should not be allowed, because all patches have to go through a testing and approval process here. If a previously approved patch gets revised in such a way that it then automatically installs on systems it was not approved for, it can do a lot of damage.
Linux gets it right: change a package, it's a new version, end of argument.
"The Linux desktop arrived in 1998 with RedHat 6.0."
Yes! Same here.
Actually it was Red Hat 5.0 which I first installed in 1997, but I managed to stuff the boot process and couldn't figure out how to install it until 1998.
It was 1998 when I started dual-booting and then 1999 I think when I went Linux cold-turkey. And it probably was RH 6 I used then.
Ah, memories.
Oh, and probably the most annoying:
7. Modes rather than options.
What makes Windows Explorer, from Win95 to XP, great is that it is customisable, on a per-user and/or per-directory basis. It has a number of orthogonal options which can be switched, and it then remembers that view. Such as: toolbars, show/hide status bar, major mode (thumbnails, list, detail etc), arrange-by, side pane (which remembers its size and is dismissable with one quick X), 'up', 'back'.
These options are great. You don't need to mess with them when your view looks ok, but since you 'live' in a file browser so much, you need to be able to tweak it when things grate.
Nautilus doesn't do this. Instead of options, it has modes. You can't pick and choose the view options *you* want for each folder, you can only pick from a tiny subset of ones the *developers* thought you *should* want.
It's little things like not being to turn off the status bar unless you're in spatial mode, not being able to adjust the size of the side pane, not being able to dismiss the side pane without hitting the menu (because usually you bring up the pane to navigate, then once you're there you need more screen real estate in a hurry) - these little, pointless restrictions just chafe.
There's no reason why the user experience needs to be restricted like this. The 'spatial' argument was where it showed the most. 'The user is using it wrong'. No. That's never the right answer.
It's a design philosophy which needs to be changed.
"Linux could clean Windows' clock if the GUI were more dependable. Right now it's pretty good, but occasionally falls flat on its face."
And please, please, PLEASE can we get (on Gnome) a DECENT replacement for Nautilus.
Nautilus just makes me wince. It's like it's taken half a squint at Windows Explorer circa 1998, on a very bad day, upside down, in a mist, and then looked at Mac Finder and did its best to forget all the good things in Explorer.
Whoever did the original Win95 Explorer design needs a medal: it's still the best feature of Windows. It's simple, intuitive, and it doesn't penalise advanced users. This is what Nautilus should be, but isn't even trying.
(Vista has also done its best to go the same route, and throw out all the things that made Win95's Explorer work.)
These are the silly things that Nautilus does:
1. Half-implemented 'web view' pane. It's useless. If I want to view something as a web page, I'll use Firefox or the browser of my choice (please not Epiphany).
2. No decent tree view. It's tacked on in the side pane, but feels ugly, horrible, restricted. It's not as easy or flexible to use as Explorer's. It's an afterthought.
3. No simple TEXT BOX view of the current location - or if it's available, it's hidden. OSX and Vista have both abandoned this but that's no reason for Linux to. The location needs to be a text box so you can Copy/Paste. That's important for advanced users, because locations are not opaque things that you can 'discover' through a conversation process, but are things you need to *communicate* to other programs and to humans. Text is the only reliable way of communicating, icons don't cut it (you can't cut and paste an icon into an email or IM or config file).
4. 'Emblems'. Sort of cute idea, but implementing anything like this at the gui file-browser level is the Wrong Place to do it. Again, because you can't communicate the presence of emblems - it's metadata that only exists in an interactive browsing session. So you can't share emblems, you can't copy/paste them, then don't exist for anyone but you and only when you're using Nautilus. So useless.
5. No decent 'detail view'. Zoomable thumbnails are sort of okay (though it's very slow to process thumbnails when you're copying a bunch of photographs), but sometimes you really do need to do some serious forensics on a directory and instead of having to drop into command-line, it would be nice to have a somewhat pleasant GUI view of the real files that are there without trying to talk down to you. Nautilus keeps trying to belittle the user and hide them from information 'for their own good'. It's a bad Apple habit, and Windows (pre-Vista) learned not to do it. Stop it.
6. 'Spatial mode'. Nuff said. No, it wasn't innovative, nor was it pleasant. Win95's Explorer had this - as one of two modes that you could select, and advanced users quickly found 'open in same window' much more usable.
Thank goodness Ubuntu hacked it off and made Nautilus nearly usable, but the Gnome folks' response still leaves a nasty taste in my mouth.
"Ultimately there's not $20, there's the $10 I used to have but instead gave to you, and at the end of the day I can at best expect to end up with that $10 back plus interest."
Something I've always wondered: how, exactly, are you going to get that $10+interest back unless at least enough new money to cover the interest is actually *created* rather than just *loaned out*?
Isn't the whole concept of interest-bearing debt really a Ponzi scheme? Eventually it's going to fall over because you can't keep loaning money that doesn't exist and then asking for more back than exists.
Why don't we just forgive all the debts and start again?
"I too disagree with Steven's argument. But people who jump on "tubes" often do not even know the concepts behind the analogy. In a lot of cases, the people that laugh at his comment are even less informed about the topic than Stevens."
Yes! THANK YOU!
I get really annoyed with Slashdot people who laugh at 'tubes' yet still use the word 'pipe'.
More importantly, the Internet infrastructure *does* have non-infinite capacity, and filesharing of movie-sized files, legal or not, *does* cause non-zero consumption of that capacity.
Internet access *doesn't* come for free or even cheap, and a flat monthly access fee *doesn't* necessarily scale to cover all conceivable maximum-bandwidth uses.
In New Zealand and Australia, we pay for our monthly data transfer usage. It's fair, it's simple, and everyone's happy. I don't understand why Americans can't cope with the concept of 'non-infinite supply of a commodity' and 'paying a fair price for what you consume'.
Seriously, the rest of the world looks at you guys sometimes and goes 'huh?'
"Everybody and his uncle tries to make systems that will index every piece of crap on your PC and it invariably results in a useless and horrible waste of resources."
On the contrary, we should seriously be asking ourselves *why*, when all our data is sitting there on our PCs, we've let ourselves get into such a state of disorganisation at the operating system level that a class of program called 'indexer' exists as a third-party tool in the first place.
How come it's not already taken as given that the primary thing an operating system *does* is, you know, *know where all its data is*?
It's as if we're living in an age before 'directories' were invented - or before databases had 'indexes' and 'queries' - and we have to manually write down and key in raw sector numbers every time we open a file. And we're okay with that, because we think - and teach - that that's 'just how computers work'. We've accepted that there's a whole class of things our computers can't do 'because there's no application to do that'.
Something is wrong with this picture.
"I think that is where something like Nepomuk could succeed where internet-wide semantic standards fail to gain traction. People are lazy but they DO devote more effort to organising their own personal data vs. what is on a web app. I do make more effort to tag photos with metadata in F-spot for example. If there is a "structural-level" standard that could be applied to all files so I can tag spreadsheets, photos, databases, addressbook contacts...whatever..and I could follow one simple, consistent process to apply my own metadata and search and organise on that data then it is very useful."
Yes, exactly. I have been wanting something like this for several years now.
I'm sick of how our desktop applications still organise data into little silos that don't communicate. All this data - photos, MP3s, email, IMs, Web bookmarks, text documents - has a fairly rigid structure, or well-defined metadata. So representing it as RDF presents no major semantic problems - the schema's already been defined.
The main problem is infrastructural: we still have this belief that we need to 'run an application' to access a piece of data. I have to 'launch my mailer' to read an email; I have to launch my browser to read a web page; I have to launch Word to read a Word document, and so on. Even if I don't want the whole GUI experience and just want to correlate some metadata. I don't even have a way to *talk* about correlating metadata between Word and my MP3 collection.
We've somehow linked *knowledge* to *human-initiated, application-centred tasks* -- and I blame object oriented programming for spreading this idea by binding 'code' and 'data' together -- but the two are nowhere near the same, and this linkage really limits what I can delegate to my computer.
An MP3 *song* is not an instance of an MP3 *player*, a Word *document* is not an instance of the Word GUI. Nouns and verbs should *not* be tightly bound together. I should be able to analyse the files (nouns) in my system completely independently of the 'user interfaces' (verbs) I can use to manipulate them. But so far, I can't.
Some kind of desktop-wide data framework that can let me construct queries between the structured *data* represented by my emails, my MP3s, my calendar and my photos, for instance, would let me create whole new forms of 'application' or 'applet' on the fly. And blur the distinction between 'user' and 'application developer' in the same way that creating SQL queries blurs the distinction between 'database user' and 'database administrator'.
I've used database systems where users can't construct ad-hoc queries - they're painful, just like today's Windows desktops.
We never should have separated those roles in the first place, and if RDF can take us back to letting *me* decide how to use and interpret *my data*, that's a step forward.
"Regardless of cause, given the failure of semantic markup in such a limited and controlled scope, it is very unlikely that it will succeed with the richness and complexity of all the data in the world."
What the heck is *with* this argument? It's completely false.
Look, we already have exactly the same thing as what the Semantic Web is trying to do: it's called SQL. Oh noes! Trying to describe real world information on a computer will fail! It's impossible!
And yet, impossible or not, we have databases. And your desktop is full of semantic markup.
Your digital camera, that takes photos, did you know it does 'semantic markup'? But it does. EXIF data attaches time, date, shutter settings to your JPG. Your MP3 collection has 'semantic markup' - the ID3 tags show title, author, album. Your email has 'semantic markup' in its headers and in the folder that you sort it into. Your calendar has 'semantic markup' in its time and date fields; if you have iCalendar, you can publish this to the Web.
Wikipedia is full of semantic markup, in its 'category' tags and links and in all those little infoboxes on each page. There's now a Semantic MediaWiki fork which uses RDF to store that data so you can query it.
The Semantic Web is nothing but an attempt to *standardise data representation* so you can do SQL-type queries between data sources you trust but might not reside on your machine: your photo collection, your MP3s, your friends' calendars. It's nothing more magical than web-distributed RDF, which is nothing but SQL restricted to triples, ie, tables of two columns. And we already understand SQL.
Quit the philosophical strawmen and realise that what you claim is impossible is *already happening*.
The reason we don't yet have *massive* RDF adoption is that it takes a while for people to catch on to the kind of crosslinking that can be done between applications; we've been trained by Windows to believe that linking data and creating mashups is *hard* and should only be done by experts. But things like Nepomuk are showing slowly how it can be done.
Right. There seems to be a huge amount of (almost wilful) misunderstanding about what the Semantic Web is trying to achieve.
I think the simplest way to describe it is 'networked databases'.
We already understand databases, and don't have any philosophical problems with 'but what does the data MEAN' and 'but what if it's WRONG'.
Why is it so hard to believe that we could advertise data on the Web in a similar way? And then link small, trusted data islands up into larger webs?
That's all the Semantic Web is trying to do: put data on the Web. Put like that, it seems obvious, doesn't it?
Or Garibaldi, for that matter.
"We just need 300 Spartans to man the bottleneck link between here and China."
And a bunch of Warthogs and Pelicans, and if we could get a couple of Arbiters with energy swords...
"I would say something like "this is the way the world ends""
Not with a bang but a headcrab?
"The media was already questioning the war long before the WMD claims were debunked. "
Yes, because it was obvious to anyone paying attention long before the war that the WMD claims were outrageous lies. The US intelligence outfits were leaking like a sieve trying to get people to notice the truth. If you followed a war-centric news aggregator like, for instance, antiwar.com (run by Libertarian types, not leftists btw) in late 2002 or early 2003, you could see literally dozens of reports coming out from various highly-placed sources frantically trying to defuse the war before it exploded. Comparing the quality of the information side by side, there was no comparison.
Usually it was small media outlets, local newspapers and the like, which would pick up a quality anti-war story from a top source. Then it would die quietly before it reached the 'big boys', the New York Timeses or the Washington Posts. The amount of coverage the big boys gave to a story seemed inversely proportional to its importance; weird tiny semantic things like the Plame case exploded into big scandals because they weren't directly criticising the *war*, just specific tactics for fighting it.
Watching the whole thing play out actually gave me *more* faith in American democracy. It revealed that there wasn't a big conspiracy controlling everything, because they sure couldn't stop the truth from getting out. There was only a loose network of small conspiracies. All they could do was delay the big media outfits from questioning the 'facts' until it was too late.