doomed to failure until someone implements something like the Dewey Decimal System for web pages
Yes, we're stuffed -- but Dewey Decimal isn't the answer (we can do a lot better than that).
There's an initiative around that's gaining considerable momentum - the Semantic Web. It starts from one bright idea by one guy, but as the guy in question is Tim B-L, then he gets listened to. There are solutions to all this. We've barely started on what we could easily achieve for indexing the web, without even trying for the really hard stuff.
Once basic semantic level indexing becoms commonplace, through tools like Dublin Core, then take a look at ontological descriptions and projects like DAML.
There's a huge amount happening in this field research-wise, it just hasn't hit the punter's web yet.
Re:MacOS X OK; 1-Button Mouse Sucks; Apple Stagnat
on
Ever Improving Laptop
·
· Score: 1
buy whatever freakin' mouse you want, if it is USB the mac will support it
if the DoubleClick (or TIVO, whatever) data is actually *worth* something [...] how come DoubleClick is on the verge of bankruptcy?
That's a good point, but it confuses value (market value) with value (cost to me for losing the privacy). Being "hard to lose" might lead to someone investing in DoubleClick, but it doesn;t mean that this stuff is really worth anything.
DoubleClick can rot in hell and I'll dance on their grave for I care. For those who work there and read this, WHY ? What do you do for an encore ? Sell crack to your kid sister ?
You can't sell porn to Slashdot readers (they're too cheap) because the 50% who aren't your target audience will get really pissed over it. You still need the targetting, even if you know the average is good.
Face it, your individual viewing data is WORTHLESS. You're just not that important of a person.
I disagree. Suppose I claimed "Your doubleclick cookies are worthless". This is true, provided there's no way to aggregate them further, but becomes very suspect information once there's a way to aggregate them further.
Suppose the TiVo starts collecting data about narrow-target pay-per-view cable channels that I subscribe to ? Mining the data that many late-night subscribers to "Dwarves in Very Large Boots" are also watching the religious guilt flagellation show at family prime time is exactly the sort of data synthesis that raises issues here.
What happens with an expansion of interactive TV a few years hence (as is very likely here in the UK) ? It's almost certain that TiVo's here will soon be in a position to see that I watch the interactive home shopping channels for hours a day, and which items I watched the most, and that I watch the all-state tractor pulls on the sport channel. The extra opportunities for not only synthesis, but for identifying me from my purchases are very close to the problems with DoubleClick (the ability to dis-anonymise many sites as soon as they've tracked me on a single non-anonymous one).
I trust TiVo (meaning literally what it says - I believe their honesty and intention to not sell the data). I have no problem with what TiVo have done. OTOH, I will remain receptive to news of them changing policy (Toysmart ?) and I'm certainly suspicious in the general case of other companies with similar tech.
Re:Bluetooth - necessary in 802.11 world?
on
Bluetooth Bombs
·
· Score: 1
If you are in france 802.11 is illigal because of the freq. it uses.
Fine. We'll use it for all the eBay nazi regalia listings.
I don't understand BlueTooth. I see how it works as a protocol for un-wiring the various devices I own personally. If I want my fridge to talk to my phone, so that it can order more milk on demand, then I can see how to do it. If I want a cellphone that makes me look like Lt. Uhura, and can play MP3's from my PDA over the same earpiece, then I can see how to do that.
What I don't understand is how I use it to do the things I currently do with a Palm and IR. How do I walk into a First Tuesday (sic) meet and beam my business card to one person, in such a way that they trust to receive it, and I don't simultaneously broadcast it to the entire room ? Despite some searching, I can't find a way to do this securely and reliably with BlueTooth, in a manner that allows ad hoc communication between devices who interact only once and fleetingly.
Is BlueTooth really so limited that it can only be useful for "my stuff" that I have previously spent time giving personal introductions to each other ? This seems like they've missed the big picture on usability in a big, big way.
I just wish HP JetSend had seen some wider adoption. It was a staggeringly clever protocol that solved many of these "surface based computing" problems you mention. Despite the limited scope of the few use cases we did see, it wasn't just a protocol for printing over IR from handhelds.
Wrong. Actually there was CCTV footage of the players and their targets until just before the incident, and of the players moving away afterwards
Why is that wrong ? The court case is relying on the eye-witness evidence, not the cameras -- because the cameras have proven to be useless at actually solving this sort of crime. It took place nearby to a camera, yet even then this piece of expensive and intrusive hardware hasn't achieved a conviction.
The "ring of steel" has been a real triumph of police surveillance cameras. Without its intel, the Met police would never have found out that most of the IRA's active service units in London turned out to be black.
due to a high profile and sustained terrorist bombing campaign in England & Ireland
There hasn't been a "high profile and sustained" campaign in either England or Ireland. There was a sustained campaign in England in the '70's (before widespread cameras), but nothing comparable since. A few(large) incidents are nothing compared to how it was 20-25 years ago.
Ireland isn't the country you think it is. You mean Northern Ireland, where incident rates are still higher then England (although also a fraction of what they were over a decade ago). Despite this, NI has fewer cameras in city centres than England, and hardly any in some major trouble spots (they'd be seen as too provocative).
This is a good thing, because it has worked in keeping levels of crime in our cities down,
Which cities are those ?
Have you visited an urban sink-estate lately ? Not the posh shopping centre of town, but the poor end, where there's no budget for cameras. Friday night violence has certainly moved, but it's not reduced. Seen Newport, or much of South Yorkshire, when the pubs close ?
CCTV footage has led to convictions for many people committing acts of violence,
Examples please. James Bulger ? Filmed on video being led away for murder, but it didn't stop it. The current case, where a number of (apparently) famous footballers severely beat and kicked one man, yet the witnesses in court are all eye-witnesses, not camera opearators.
Never, ever, accept a new position until you've met all your managers/supervisors,
I should have known better. An MCSP site, offering to lobotomize me into being an MCSD. I resigned at the end of the 2nd day, after discovering the control-freak "technical" director grew up writing Pick, and still thought in it.
Third day was a pre-arranged site visit, so professionalism held me to it (suit and all). Their much vaunted split-site Access database with batch syncing by email updates turned out to be an impressive piece of design, spoilt only by them having mapped the file paths between sites onto databases stored at the other site (Doh!). Not surprisingly they were suffering poor performance and corruption problems when the line went flakey.
Checking today, I see their homepage has the traditional greeting, "We have noticed that you do not have the Latest Flash plugin. The <foo> web site, has been built using Flash technology, and you will need to download it to enter this part of the site.". I'm so glad I left there...
The Guardian [...] is seen as a trendy middle-class pseudo-intellectual broadsheet,
The Grauniad is reasonably objective, but the ICA is gaining an increasingly bad reputation for black-turtlenecked post-modern wankerdom from the technically illiterate chattering classes.
I think that cars are a good analogy, but not in the way that you describe.
My car doesn't go wrong. To quote Coupland, "some time in the late '80s cars got a lot better" (Girlfriend in a Coma). I used to spend a long time working on my several cars. These days I no longer have the time or quite the same enthusiasm but fortunately I no longer need it. Cars just don't fall apart anything like they used to....and it's certainly not that I spend any more on them than I used to 15 years ago.
Computers are the same. My Mother got her first PC this year. It has an external SCSI interface to a film scanner. Now when I originally bought that card, some years back, I spent a hellish Summer sorting out SCSI and driver problems on a roomful of scanners. This time I talked my Dad through it over the phone. Scream at Bill all you like, but even Windows just doesn't suck like it used to.
It's no longer necessary to understand NS4's broken HTML rendering to make a useful web site. Just use CSS - there's enough good support for it now, on every OS platform, that relying on CSS is no longer the brave but dumb choice it used to be. One day, even FrontPage will work right.
This stuff gets easier every day. OK, so there will always be a Digerati, but they'll be the ones fooling with their VR goggles and optic-nerve implants in the future. Granny will be writing HTML with a rented and copy-protected version of BillWriter, and not thinking twice about it.
Will there be a social divide to the Digerati ? I don't think so. The big risk was that the non-Digerati would be disadvantaged by an inability to learn / find the best bargain / partake in enjoyable recreations, by not having net access (Actually, the more I think of it, the less vital the Web seems). They'd also lose out on the job market, because knowledge workers needed to have practiced their 'Net skills beforehand to land the best jobs. Well I'm sorry, but some dumb Johnny-No-Stars doesn't become smarter, just because they've got a WebTeeVee in their trailer. Access to the tools is important, but so is having something to offer when you use them. If you're smart enough to contribute with them, and you're born into the affluent West, then it's no great stretch to gain the access.
A friend of mine recently toured some of the poorer parts of Brazil, with a particular interest in virtual communities and 'Net access from barrios. One of her findings was that there was often the kit in place, and even the power to run it, but it wasn't being used owing to simple problems like illiteracy. There's always a social divide, but the "digerati effect" isn't causing it -- it's just called poverty.
Absolutely. But I think it's no less a robot for that.
What is a robot ? Are there any existing devices that you'd class as a robot ? I think this is an interesting question, because there are very, very few current machines that meet your criteria.
No, a machine has to be able to make decisions in order to qualify as a robot.
What's a "decision" ? Does a robot that "makes a choice" that will improve its access to food/fuel/power count as a decision maker ? Must it have an internal world model that knows, "I want food, Food is over there, I will go over there" ? Or is it sufficient to simply turn towards the light ? It may not have an understanding of "food", "light" and their relationship, merely that generations of its parent [algorithms] underwent a proces of natural selection that rewarded those who favoured light by improving their food access.
Are you familiar with subsumption architectures ? These are a highly successful series of robot designs, based on tiny robot-components executing according to their inbuilt rules, but producing a resultant behaviour of the system (or "creature") that is apparently far more complex. Nowhere in these architectures is there any notion of a Big Architecture "World View" or any "Reasoning"
One of the dogmatic statements of the subsumptionists (on Brooks' model) is that, "The world is its own best model". This means that robots shouldn't try to model the world, they should use it itself as the model. For a photovore, the best food sources are found in the places that are brightest (this is a world modelling task, as it's not trivially obvious to a robot). A Big Architecture robot would need to "know" that light==food, and that it should then seek out food by seeking out light. A photovore doesn't do this; it just goes to the brightest places, with all the intelligence of a thermostat. It doesn't even realise it's seeking "food" by this process, it only knows that it's seeking light. It doesn't even know why it seeks light (as it doesn't know that light is food), it just does it. More complex ones may have built this behaviour up randomly, by either a reward or selection process; perhaps the sound and humidity seekers starved earlier.
A robot is a machine that can adapt its behavior to achieve its goals.
That's an interesting distinction to make, and clearly a valuable one in some cases, but it's certainly not a fundamental definition of a robot. Even Kapek's original robotniks didn't have that much freedom of behaviour. Few robots even have the concept of goal-oriented behaviour, let alone choosing behaviours to achieve it. Much of the interesting current work in robotics is focussed on robots built as nodes with minimal inherent behaviours and the behaviour of the system as a whole arises as their combination, not by designing from a top-down concept.
Radio just isn't going to cut it for this. Simple and reliable 'bot positioning at this time needs an absolute position indicator. Open the box, place the 'bot on-site, boot it up for the very first time and it should be able to work out where it is. Anything that requires it to be placed in a "known position", then track movement from their is just too painful to contemplate.
Doppler radio systems might deliver some velocity information (but it's difficult), but it cannot do absolute positioning.
Radio multi-station phase-tracking systems were developed during WW2 (Gee, Oboe) to give absolute positions. The trouble with these is that they only work over long distances. They work by timing path length differences between fixed stations, so they can only work if the path lengths differ by more than the shortest time interval you can measure, multiplied by the speed of light. With radio wavelengths in a garden - forget it !
Slashdot Mirror
doomed to failure until someone implements something like the Dewey Decimal System for web pages
Yes, we're stuffed -- but Dewey Decimal isn't the answer (we can do a lot better than that).
There's an initiative around that's gaining considerable momentum - the Semantic Web. It starts from one bright idea by one guy, but as the guy in question is Tim B-L, then he gets listened to. There are solutions to all this. We've barely started on what we could easily achieve for indexing the web, without even trying for the really hard stuff.
Once basic semantic level indexing becoms commonplace, through tools like Dublin Core, then take a look at ontological descriptions and projects like DAML.
There's a huge amount happening in this field research-wise, it just hasn't hit the punter's web yet.
buy whatever freakin' mouse you want, if it is USB the mac will support it
Tricky on a laptop though.
So, it's just like the Fujitsu, except that I can already order the Fujis ?
Why is this thing so exciting ? Are Slashdot doing paid product plugs now ? 8-(
It's cute, but it's still a Mac.
Anyone here know what the current status of BSD on the StealthBarbie is ? Soon as it runs a real OS, I'm up for one 8-)
if the DoubleClick (or TIVO, whatever) data is actually *worth* something [...] how come DoubleClick is on the verge of bankruptcy?
That's a good point, but it confuses value (market value) with value (cost to me for losing the privacy). Being "hard to lose" might lead to someone investing in DoubleClick, but it doesn;t mean that this stuff is really worth anything.
DoubleClick can rot in hell and I'll dance on their grave for I care. For those who work there and read this, WHY ? What do you do for an encore ? Sell crack to your kid sister ?
You can't sell porn to Slashdot readers (they're too cheap) because the 50% who aren't your target audience will get really pissed over it. You still need the targetting, even if you know the average is good.
Face it, your individual viewing data is WORTHLESS. You're just not that important of a person.
I disagree. Suppose I claimed "Your doubleclick cookies are worthless". This is true, provided there's no way to aggregate them further, but becomes very suspect information once there's a way to aggregate them further.
Suppose the TiVo starts collecting data about narrow-target pay-per-view cable channels that I subscribe to ? Mining the data that many late-night subscribers to "Dwarves in Very Large Boots" are also watching the religious guilt flagellation show at family prime time is exactly the sort of data synthesis that raises issues here.
What happens with an expansion of interactive TV a few years hence (as is very likely here in the UK) ? It's almost certain that TiVo's here will soon be in a position to see that I watch the interactive home shopping channels for hours a day, and which items I watched the most, and that I watch the all-state tractor pulls on the sport channel. The extra opportunities for not only synthesis, but for identifying me from my purchases are very close to the problems with DoubleClick (the ability to dis-anonymise many sites as soon as they've tracked me on a single non-anonymous one).
I trust TiVo (meaning literally what it says - I believe their honesty and intention to not sell the data). I have no problem with what TiVo have done. OTOH, I will remain receptive to news of them changing policy (Toysmart ?) and I'm certainly suspicious in the general case of other companies with similar tech.
If you are in france 802.11 is illigal because of the freq. it uses.
Fine. We'll use it for all the eBay nazi regalia listings.
I don't understand BlueTooth. I see how it works as a protocol for un-wiring the various devices I own personally. If I want my fridge to talk to my phone, so that it can order more milk on demand, then I can see how to do it. If I want a cellphone that makes me look like Lt. Uhura, and can play MP3's from my PDA over the same earpiece, then I can see how to do that.
What I don't understand is how I use it to do the things I currently do with a Palm and IR. How do I walk into a First Tuesday (sic) meet and beam my business card to one person, in such a way that they trust to receive it, and I don't simultaneously broadcast it to the entire room ? Despite some searching, I can't find a way to do this securely and reliably with BlueTooth, in a manner that allows ad hoc communication between devices who interact only once and fleetingly.
Is BlueTooth really so limited that it can only be useful for "my stuff" that I have previously spent time giving personal introductions to each other ? This seems like they've missed the big picture on usability in a big, big way.
I just wish HP JetSend had seen some wider adoption. It was a staggeringly clever protocol that solved many of these "surface based computing" problems you mention. Despite the limited scope of the few use cases we did see, it wasn't just a protocol for printing over IR from handhelds.
Handspring have realized [..] looked clunky and old-school compared to the Palm V range.
So how do you explain Sony's Clie ? That's plug-ugly, even if Memory Stick is the only really useful Palm add-on around.
Wrong. Actually there was CCTV footage of the players and their targets until just before the incident, and of the players moving away afterwards
Why is that wrong ? The court case is relying on the eye-witness evidence, not the cameras -- because the cameras have proven to be useless at actually solving this sort of crime. It took place nearby to a camera, yet even then this piece of expensive and intrusive hardware hasn't achieved a conviction.
especailly in the "ring of steel"
The "ring of steel" has been a real triumph of police surveillance cameras. Without its intel, the Met police would never have found out that most of the IRA's active service units in London turned out to be black.
due to a high profile and sustained terrorist bombing campaign in England & Ireland
There hasn't been a "high profile and sustained" campaign in either England or Ireland. There was a sustained campaign in England in the '70's (before widespread cameras), but nothing comparable since. A few(large) incidents are nothing compared to how it was 20-25 years ago.
Ireland isn't the country you think it is. You mean Northern Ireland, where incident rates are still higher then England (although also a fraction of what they were over a decade ago). Despite this, NI has fewer cameras in city centres than England, and hardly any in some major trouble spots (they'd be seen as too provocative).
This is a good thing, because it has worked in keeping levels of crime in our cities down,
Which cities are those ?
Have you visited an urban sink-estate lately ? Not the posh shopping centre of town, but the poor end, where there's no budget for cameras. Friday night violence has certainly moved, but it's not reduced. Seen Newport, or much of South Yorkshire, when the pubs close ?
CCTV footage has led to convictions for many people committing acts of violence,
Examples please. James Bulger ? Filmed on video being led away for murder, but it didn't stop it. The current case, where a number of (apparently) famous footballers severely beat and kicked one man, yet the witnesses in court are all eye-witnesses, not camera opearators.
Never, ever, accept a new position until you've met all your managers/supervisors,
I should have known better. An MCSP site, offering to lobotomize me into being an MCSD. I resigned at the end of the 2nd day, after discovering the control-freak "technical" director grew up writing Pick, and still thought in it.
Third day was a pre-arranged site visit, so professionalism held me to it (suit and all). Their much vaunted split-site Access database with batch syncing by email updates turned out to be an impressive piece of design, spoilt only by them having mapped the file paths between sites onto databases stored at the other site (Doh!). Not surprisingly they were suffering poor performance and corruption problems when the line went flakey.
Checking today, I see their homepage has the traditional greeting, "We have noticed that you do not have the Latest Flash plugin. The <foo> web site, has been built using Flash technology, and you will need to download it to enter this part of the site.". I'm so glad I left there...
The Guardian [...] is seen as a trendy middle-class pseudo-intellectual broadsheet,
The Grauniad is reasonably objective, but the ICA is gaining an increasingly bad reputation for black-turtlenecked post-modern wankerdom from the technically illiterate chattering classes.
I think that cars are a good analogy, but not in the way that you describe.
My car doesn't go wrong. To quote Coupland, "some time in the late '80s cars got a lot better" (Girlfriend in a Coma). I used to spend a long time working on my several cars. These days I no longer have the time or quite the same enthusiasm but fortunately I no longer need it. Cars just don't fall apart anything like they used to. ...and it's certainly not that I spend any more on them than I used to 15 years ago.
Computers are the same. My Mother got her first PC this year. It has an external SCSI interface to a film scanner. Now when I originally bought that card, some years back, I spent a hellish Summer sorting out SCSI and driver problems on a roomful of scanners. This time I talked my Dad through it over the phone. Scream at Bill all you like, but even Windows just doesn't suck like it used to.
It's no longer necessary to understand NS4's broken HTML rendering to make a useful web site. Just use CSS - there's enough good support for it now, on every OS platform, that relying on CSS is no longer the brave but dumb choice it used to be. One day, even FrontPage will work right.
This stuff gets easier every day. OK, so there will always be a Digerati, but they'll be the ones fooling with their VR goggles and optic-nerve implants in the future. Granny will be writing HTML with a rented and copy-protected version of BillWriter, and not thinking twice about it.
Will there be a social divide to the Digerati ? I don't think so. The big risk was that the non-Digerati would be disadvantaged by an inability to learn / find the best bargain / partake in enjoyable recreations, by not having net access (Actually, the more I think of it, the less vital the Web seems). They'd also lose out on the job market, because knowledge workers needed to have practiced their 'Net skills beforehand to land the best jobs. Well I'm sorry, but some dumb Johnny-No-Stars doesn't become smarter, just because they've got a WebTeeVee in their trailer. Access to the tools is important, but so is having something to offer when you use them. If you're smart enough to contribute with them, and you're born into the affluent West, then it's no great stretch to gain the access.
A friend of mine recently toured some of the poorer parts of Brazil, with a particular interest in virtual communities and 'Net access from barrios. One of her findings was that there was often the kit in place, and even the power to run it, but it wasn't being used owing to simple problems like illiteracy. There's always a social divide, but the "digerati effect" isn't causing it -- it's just called poverty.
Imagine your local server room. Don't you think they would like a 20% decrese in their power bill?
We can afford the power bill easily enough. What we can't deal with is the extra cooling.
Does anyone know roughly:
It was also instrumental in spawning Open Source
That's like saying Nazi Germany spawned Israel.
I learned to rollerblade between the (softly-padded) cube walls of Lucent
8-)
it is merely a feedback controlled motor system.
Absolutely. But I think it's no less a robot for that.
What is a robot ? Are there any existing devices that you'd class as a robot ? I think this is an interesting question, because there are very, very few current machines that meet your criteria.
No, a machine has to be able to make decisions in order to qualify as a robot.
What's a "decision" ? Does a robot that "makes a choice" that will improve its access to food/fuel/power count as a decision maker ? Must it have an internal world model that knows, "I want food, Food is over there, I will go over there" ? Or is it sufficient to simply turn towards the light ? It may not have an understanding of "food", "light" and their relationship, merely that generations of its parent [algorithms] underwent a proces of natural selection that rewarded those who favoured light by improving their food access.
Are you familiar with subsumption architectures ? These are a highly successful series of robot designs, based on tiny robot-components executing according to their inbuilt rules, but producing a resultant behaviour of the system (or "creature") that is apparently far more complex. Nowhere in these architectures is there any notion of a Big Architecture "World View" or any "Reasoning"
One of the dogmatic statements of the subsumptionists (on Brooks' model) is that, "The world is its own best model". This means that robots shouldn't try to model the world, they should use it itself as the model. For a photovore, the best food sources are found in the places that are brightest (this is a world modelling task, as it's not trivially obvious to a robot). A Big Architecture robot would need to "know" that light==food, and that it should then seek out food by seeking out light. A photovore doesn't do this; it just goes to the brightest places, with all the intelligence of a thermostat. It doesn't even realise it's seeking "food" by this process, it only knows that it's seeking light. It doesn't even know why it seeks light (as it doesn't know that light is food), it just does it. More complex ones may have built this behaviour up randomly, by either a reward or selection process; perhaps the sound and humidity seekers starved earlier.
A robot is a machine that can adapt its behavior to achieve its goals.
That's an interesting distinction to make, and clearly a valuable one in some cases, but it's certainly not a fundamental definition of a robot. Even Kapek's original robotniks didn't have that much freedom of behaviour. Few robots even have the concept of goal-oriented behaviour, let alone choosing behaviours to achieve it. Much of the interesting current work in robotics is focussed on robots built as nodes with minimal inherent behaviours and the behaviour of the system as a whole arises as their combination, not by designing from a top-down concept.
Radio just isn't going to cut it for this. Simple and reliable 'bot positioning at this time needs an absolute position indicator. Open the box, place the 'bot on-site, boot it up for the very first time and it should be able to work out where it is. Anything that requires it to be placed in a "known position", then track movement from their is just too painful to contemplate.
Doppler radio systems might deliver some velocity information (but it's difficult), but it cannot do absolute positioning.
Radio multi-station phase-tracking systems were developed during WW2 (Gee, Oboe) to give absolute positions. The trouble with these is that they only work over long distances. They work by timing path length differences between fixed stations, so they can only work if the path lengths differ by more than the shortest time interval you can measure, multiplied by the speed of light. With radio wavelengths in a garden - forget it !
instead of shocking your lawnmower
Nah, shock the little tin bastard. Those 'bots will be getting uppity enough before too long. Best keep them in their place now.