So far there are two tracks (pardon the pun) to this discussion:
(i) Watermarks can be filtered out without serious damage to the content. I am not convinced either way.
(ii) That if they do, then it requires person specific keys for each recording - and the implications that has for the continued existence of CD's and for privacy of what you are doing.
So, is (i) true or false ? (ii) Do you think "the Man" could get away with this ?
I think thats right (easy to copy from audio channel) - However, watermarking is slightly different to copy-protection encryption. Basically they alter tiny bits of every piece of music in such a way that when it is held up to the light, so to speak, you can see the watermark.
This does not stop you from copying the song - but generally you cannot find out which bits were altered very easily (maybe Im wrong). Thus when you buy a song and copy it, your finger print will remain on the song, and if it appears on Napster, you will be fingered as the person who allowed the copyrighting...
Obviously you can also do copy-protection based on this as well, but as it is so easy to circumnavigate, it is more likely that the watermark fingerprint would be used.
Try getting Oracle 8 ODBC drivers for the Mac today....? Oracle used to produced Oracle 7 for Mac, and whilst it still works (I have a copy somewhere), it is impossible to obtain any more. However even this refused to co-operate with ODBC drivers that were available (intersolv and the other one I forget). The least you'd expect is Oracle 8 drivers so that Mac users could talk to Oracle servers.. it'd be even better if they issued the matching ODBC drivers, but no... Just a pet peeve of mine:) I rate Oracle slightly below that of Microsoft sometimes depending on how I am feeling... Winton
I thought I saw a reference earlier this week to a company that is bringing out an X server... so in theory one could just use X on OS X (man is that gonna get confusing, worse than the client/server naming in X:)).
Personally I can't wait. I love working on my Mac but you can't beat Unix for flexibility. Now we get both integrated.
My main concern though is getting PPC compiled binaries. For example, CMU Common Lisp does not have a PPC version. Maybe this is just because it is a compiler issue - but do Oracle have plans to release Oracle 8 i for the PPC platform ?
Two things - a student called Thai Tran at Stanford put together such an environment called www.booksmart.com. Seems to be dead now, you might find him using some directory services. Secondly, there is a vendor called MUZE, that can provide all the data you need (www.muze.com). I suspect they are expensive
The M in all of these stands for "markup". As in document markup - not knowledge representation.
That is primarily where it is overblown, that it will provide application/domain interoperability.
Advantages:
(i) Yes, it provides a common syntax for delimiting data, albeit verbosely. (ii) There are common tools for browsing and editting.
Disadvantages:
(i) Highly verbose (perhaps the advantage as well) (ii) Highly prone to abuse (eg XML with not DTDs anyone?) and to divergence. (iii) Differences attribute and tagged values is highly unclear.
But mainly it is not a panacea at all for information integration, which is where it is mainly being hyped.
As for Microsoft using it. Well, unless they provide synatic definitions and semantic mappings they might as well have used a proprietary syntax.
The only reason it has taken off is because people think "Gee, it looks like HTML". And when will people wake up to the fact that seeing <PRICE>20</PRICE> tells you little...:) The worst thing about XML is that there is no consistent data/object scheme, even though this should have been a breeze.
and lots of other combos:) Assuming you can agree on what PRICE even is (eg includes Sales Tax etc).
I guess it is reasonable (and heavily verbose) to mark up with XML... but it really is kind of sad to see it being such a big hit, when much better thought out interoperability schemes get no press (such as KIF, KQML, FIPA).
All the effort in XML seems directed towards making it work better in the directions of querying/relatiting concepts.
XML is fine for CSS.... and maybe for making previously cryptic configuration files more editable. But think about MacOS X - Are the XML files really going to be any more easily understood than the old UNIX/X config files ?
Widget.window.x = 10 Widget.window.y = 10
versus
<WIDGET><WINDOW><X>10</X></WINDOW></WIDGET>
Nothing against XML, one just has to ask, what does it really buy me ?
I totally agree - Neal, if you're listening, please explain why your novels are great down to the last chapter or two. The endings are always kind of non-conclusive/too formulaic.
I can't for the life of me explain it, but I just read Zodiac (first book last in my case), and the ending was just a let down.
Maybe its because he never writes series about the same characters... and maybe writes them in that frame of mind.
Doesn't invalidate his books, Cryptonomicon was a superb work.
Winton
Doh, No Macintosh version either yet...
on
Diablo 2 Goes Gold
·
· Score: 1
Well I feel for you guys with Linux desktops and laptops... but Mac users will get it by third quarter 2000 (whenever that is really). On the plus side I guess I can finish my thesis first:)
Anyone though about the new Mac OS X ? or the Crusoe and its new architecture ?
Otherwise, its Nail on head time:-)
I'm just finishing my thesis (in machine learning), and what am I using to write it -- TeX and Emacs. (Anything is better than Microsloth Word...).
I'm not a Systems Research person, but I definitely have been feeling the same way - when is something truly exciting going to happen in Comp Sci ? The first burst of the web, in the early 90s was truly a "wow" time... now its just truly staid again. I think maybe I'm just a jaded AI/Software Agents researcher though:)
No experience of them personally, but I'd suggest LOGO or Mindstorms (which I think is a descendant of LOGO, at least they both emerged from MIT as teaching tools).
I learnt pretty simple problems in BASIC. I think any interpretative language would be good.
A heavy vote against Python - syntax errors because of having too many spaces....
Whatever you choose, it has to have good informative error messages and be easy to debug. I'd hesitate to suggest Java and certainly PERL. Although I love LISP, I probably wouldnt recommend it as a beginning language either!
Not wishing to throw water on this idea - but given that this isn't socialist Europe - Bill is not going to lose ownership of shares in all three companies is he - nor are all the other stockeholders. So how does this work ? Microsoft won't be nationalised...
I like the comment about Microsoft losing the thing that gives them the monopoloy power - IP protection... (copyright, trademarks, patents, trade secrets).
Traditional economics uses supply and demand to set price. Marxist economics talks about average labour time as determining the exchange rate (and profit coming from doing more efficiently than others, and/or holding capital which can get labor our of other people).
In the context of approxinately zero cost of reproduction, then all we end up with is a situation where the value is only created by artificially restricting duplication (subject to laws of monopoly?).
Even the traditional artistic source of value (uniqiueness of a performance or artifact), doesn't apply.
Is there an alternative ? We have a system, where by talented people want to give up their time to create art (or whatever it is that Metallica produces:) ), and should expect in exchange to recieve a reward in proportion to their contribution to society (in Metallica's case, perhaps Naptser is the right approach:)).
Hmm, I'm right out of ideas... what is the value of something that has zero reproduction costs in the abscence of being able to artificially restrict supply ?
ok - how about this -- once the reproducible work has been done, then the stuff is "out there". Perhaps Metallica should pre-auction their next album - "How much would you pay for this album to be produced (or not as the case maybe)". They can then decide whether it is worth their while. If they do, and they produce it for the distributed contract, then the world has a new album, which is now free to everyone. I just realised this is like that open source market place I saw recently.
Hee, I like this one - Imaging bidding to prevent the production of Barney...
I liked the comment someone made about making it from the CSM/Syndicates point of view.
Background Mulder and Scully, so they get guest roles, but basically have random FBI agents getting assigned to cases which turn out to be X-Files, but there's no X-Files office to assign weird stuff now, so CSM/Syndicate have to cover it from the initial FBI investigator(s) as well as they can...
Slashdot Mirror
So far there are two tracks (pardon the pun) to this discussion:
(i) Watermarks can be filtered out without serious damage to the content. I am not convinced either way.
(ii) That if they do, then it requires person specific keys for each recording - and the implications that has for the continued existence of CD's and for privacy of what you are doing.
So, is (i) true or false ? (ii) Do you think "the Man" could get away with this ?
Winton
moderate up maybe ? Very informative !
I think thats right (easy to copy from audio channel) - However, watermarking is slightly different to copy-protection encryption. Basically they alter tiny bits of every piece of music in such a way that when it is held up to the light, so to speak, you can see the watermark.
This does not stop you from copying the song - but generally you cannot find out which bits were altered very easily (maybe Im wrong). Thus when you buy a song and copy it, your finger print will remain on the song, and if it appears on Napster, you will be fingered as the person who allowed the copyrighting...
Obviously you can also do copy-protection based on this as well, but as it is so easy to circumnavigate, it is more likely that the watermark fingerprint would be used.
Winton
Don't look much alike to me...
Winton
Patently absurb...
Remember the British Poll Tax ? Thatcher does, that's why she's not in power any more...
Winton
Is there any relationship between Mac OS X and Robert X Cringely ? If there is, why isn't it Mac X OS ? :-) Winton
Try getting Oracle 8 ODBC drivers for the Mac today....? Oracle used to produced Oracle 7 for Mac, and whilst it still works (I have a copy somewhere), it is impossible to obtain any more. However even this refused to co-operate with ODBC drivers that were available (intersolv and the other one I forget). The least you'd expect is Oracle 8 drivers so that Mac users could talk to Oracle servers.. it'd be even better if they issued the matching ODBC drivers, but no... Just a pet peeve of mine :) I rate Oracle slightly below that of Microsoft sometimes depending on how I am feeling... Winton
I thought I saw a reference earlier this week to a company that is bringing out an X server... so in theory one could just use X on OS X (man is that gonna get confusing, worse than the client/server naming in X :)).
Personally I can't wait. I love working on my Mac but you can't beat Unix for flexibility. Now we get both integrated.
My main concern though is getting PPC compiled binaries. For example, CMU Common Lisp does not have a PPC version. Maybe this is just because it is a compiler issue - but do Oracle have plans to release Oracle 8 i for the PPC platform ?
Winton
... from a country where the average age to get a drink legally is 21...
:)
Winton
Well apart from all the issues raised so far, the one I haven't seen addressed is how many external modules are developed for it.
So, I code in LISP, and great it is too, if I want to write my own code...watch peoples faces as you build up a function they need in 5 minutes.
However, let's say I want to do some XML processing - There is a CL-HTTP package that has some stuff in it. But XSLT Support ? Forgerraboutit....
Meanwhile, my Java colleagues are happily getting on with fighting Java, rather than worrying about the XSLT coding.
This is my take on one of LISP's problem - and maybe perhaps its just descended from the small number of users.
But it does kill me how hard it is even to get some basic system calls to work in LISP, where as Perl they come built-in....
Cheers,
Winton
At least Vannevar was a real visionary, working before networks and desktop computers had existed...
Now I finally got to see the pictures, Note that:
(a) It is AppleInsider (maybe the least reliable Apple source imho)
(b) It looks like a glove without the back... I can actually imagine that squeezing ones fingers down will be kind of interesting...
Maybe like a poor persons VR glove ?
Winton
Two things - a student called Thai Tran at Stanford put together such an environment called www.booksmart.com. Seems to be dead now, you might find him using some directory services. Secondly, there is a vendor called MUZE, that can provide all the data you need (www.muze.com). I suspect they are expensive
The M in all of these stands for "markup". As in document markup - not knowledge representation.
That is primarily where it is overblown, that it will provide application/domain interoperability.
Advantages:
(i) Yes, it provides a common syntax for delimiting data, albeit verbosely.
(ii) There are common tools for browsing and editting.
Disadvantages:
(i) Highly verbose (perhaps the advantage as well)
(ii) Highly prone to abuse (eg XML with not DTDs anyone?) and to divergence.
(iii) Differences attribute and tagged values is highly unclear.
But mainly it is not a panacea at all for information integration, which is where it is mainly being hyped.
As for Microsoft using it. Well, unless they provide synatic definitions and semantic mappings they might as well have used a proprietary syntax.
Winton
Hear, hear, hear !!!
:) The worst thing about XML is that there is no consistent data/object scheme, even though this should have been a breeze.
:) Assuming you can agree on what PRICE even is (eg includes Sales Tax etc).
The only reason it has taken off is because people think "Gee, it looks like HTML". And when will people wake up to the fact that seeing <PRICE>20</PRICE> tells you little...
<PRICE UNITS="DOLLARS" VALUE="20.00"/>
<PRICE><DOLLARS>20</DOLLARS></PRICE>
and lots of other combos
I guess it is reasonable (and heavily verbose) to mark up with XML... but it really is kind of sad to see it being such a big hit, when much better thought out interoperability schemes get no press (such as KIF, KQML, FIPA).
All the effort in XML seems directed towards making it work better in the directions of querying/relatiting concepts.
XML is fine for CSS.... and maybe for making previously cryptic configuration files more editable. But think about MacOS X - Are the XML files really going to be any more easily understood than the old UNIX/X config files ?
Widget.window.x = 10
Widget.window.y = 10
versus
<WIDGET><WINDOW><X>10</X></WINDOW></WIDGET>
Nothing against XML, one just has to ask, what does it really buy me ?
Winton
This link SF Chronicle: Workers Take In Stride has an image of several Bay Area Microsoft employees sitting round an ...
...Apple Macintosh Powerbook!
:-)
I totally agree - Neal, if you're listening, please explain why your novels are great down to the last chapter or two. The endings are always kind of non-conclusive/too formulaic.
I can't for the life of me explain it, but I just read Zodiac (first book last in my case), and the ending was just a let down.
Maybe its because he never writes series about the same characters... and maybe writes them in that frame of mind.
Doesn't invalidate his books, Cryptonomicon was a superb work.
Winton
Well I feel for you guys with Linux desktops and laptops... but Mac users will get it by third quarter 2000 (whenever that is really). On the plus side I guess I can finish my thesis first :)
FWIW, Bungie is still number one in my eyes.
Winton
Anyone though about the new Mac OS X ? or the Crusoe and its new architecture ?
Otherwise, its Nail on head time :-)
I'm just finishing my thesis (in machine learning), and what am I using to write it -- TeX and Emacs. (Anything is better than Microsloth Word...).
I'm not a Systems Research person, but I definitely have been feeling the same way - when is something truly exciting going to happen in Comp Sci ? The first burst of the web, in the early 90s was truly a "wow" time... now its just truly staid again. I think maybe I'm just a jaded AI/Software Agents researcher though :)
I think it should be like the DMCA, they have to swear under liability for perjury that there is no existing prior art.
Then you'll have to be damned sure you've done your homework before you submit one.
Winton
No experience of them personally, but I'd suggest LOGO or Mindstorms (which I think is a descendant of LOGO, at least they both emerged from MIT as teaching tools).
I learnt pretty simple problems in BASIC. I think any interpretative language would be good.
A heavy vote against Python - syntax errors because of having too many spaces....
Whatever you choose, it has to have good informative error messages and be easy to debug. I'd hesitate to suggest Java and certainly PERL. Although I love LISP, I probably wouldnt recommend it as a beginning language either!
Winton
Moderate up further! Very funny!
Not wishing to throw water on this idea - but given that this isn't socialist Europe - Bill is not going to lose ownership of shares in all three companies is he - nor are all the other stockeholders. So how does this work ? Microsoft won't be nationalised...
I like the comment about Microsoft losing the thing that gives them the monopoloy power - IP protection... (copyright, trademarks, patents, trade secrets).
Winton
Traditional economics uses supply and demand to set price. Marxist economics talks about average labour time as determining the exchange rate (and profit coming from doing more efficiently than others, and/or holding capital which can get labor our of other people).
In the context of approxinately zero cost of reproduction, then all we end up with is a situation where the value is only created by artificially restricting duplication (subject to laws of monopoly?).
Even the traditional artistic source of value (uniqiueness of a performance or artifact), doesn't apply.
Is there an alternative ? We have a system, where by talented people want to give up their time to create art (or whatever it is that Metallica produces :) ), and should expect in exchange to recieve a reward in proportion to their contribution to society (in Metallica's case, perhaps Naptser is the right approach :)).
Hmm, I'm right out of ideas... what is the value of something that has zero reproduction costs in the abscence of being able to artificially restrict supply ?
ok - how about this -- once the reproducible work has been done, then the stuff is "out there". Perhaps Metallica should pre-auction their next album - "How much would you pay for this album to be produced (or not as the case maybe)". They can then decide whether it is worth their while. If they do, and they produce it for the distributed contract, then the world has a new album, which is now free to everyone. I just realised this is like that open source market place I saw recently.
Hee, I like this one - Imaging bidding to prevent the production of Barney...
Winton
I liked the comment someone made about making it from the CSM/Syndicates point of view.
Background Mulder and Scully, so they get guest roles, but basically have random FBI agents getting assigned to cases which turn out to be X-Files, but there's no X-Files office to assign weird stuff now, so CSM/Syndicate have to cover it from the initial FBI investigator(s) as well as they can...
Winton