Govt: "Give me all your URLs" ISPs: "No" Govt: "OK, will you give me half?" ISPs: "OK"
This is not a win. It is true that the government is no longer proposing new legislation. This is because they seem to be getting much of what they want without new legislation. In particular, it appears there will be a new industry "Code of Practice" which will then have force under current legislation (think of it as like changing a regulation). There is also going to be a 'police request' to the ISPs. It isn't clear to me at this point which of these, or how the combination, leads to enforceability, but it seems all Aus ISPs are now going to be filtered.
Good:
- The black-list will be the Interpol black-list, which I understand is at least reasonably well monitored and is quite small. It is also outside of political control in Australia.
- I'm guessing the filtering will be DNS based for domains on the list, forwarding them to a proxy that checks the URL. This means it will not affect people not accessing those domains.
Bad:
- We have a filter. This will have two effects: it will lead to worse enforcement of child abuse as people get complacent. It will cause the occasional problem like the Wikipedia/Scorpians cover (see http://en.wikipedia.org/wiki/Internet_Watch_Foundation_and_Wikipedia )
This is a shame. Child porn sites stay up longer than phishing sites because banks actually want the phishing sites down, not just covered up.
It will also be interesting to see what effect DNS-SEC has on this, but that is in the longer term.
Each country already has its own DNS. Country code domain names have been around for a long time. Maintaining the root servers that point to the country codes doesn't need to be an automated system. I'd hand it over to the group that agrees on the country codes: the UN.
What you're really suggesting is getting rid of non-country code domain names. All those 'blah.com' addresses would need to choose one or more country codes to occupy... 'blah.com.us' or 'blah.co.uk' or... . This would be an improvement on the current situation, but it would also be a huge upheaval and I'm not sure the long term benefits are worth the short term cost.
But even once you've done that you still haven't solved all the problems. If a UK resident purchases from a US shop, whose laws apply? US, UK or international law? It is at least clear to customer that they're purchasing from an overseas entity, but that doesn't solve the legal questions. Things could get messier still: what happens if a.co.uk domain name points to a server in the US and there is an Australian customer? Whose laws apply then?
And lets not get started on the allocation of IP addresses...:)
For every complex problem there is a solution that is simple, neat and wrong.
At 5 feet 6 inches and about 270 pounds, McArthur Wheeler is an easily recognizable man — even when wearing lemon juice on his face.
That certainly came as a surprise to Wheeler, 45, of Versailles Street, McKeesport. He was incredulous in April when Pittsburgh robbery detectives told him that he had been identified in surveillance photographs as one of the two men who robbed two banks in Brighton Heights and Swissvale on Jan. 6.
"But I wore the lemon juice. I wore the lemon juice,'' a puzzled Wheeler told the even more puzzled detectives.
The detectives' confusion turned to incredulity as Wheeler explained about his would-be lemon aid. "Someone told him that if you put lemon juice on your face it makes you invisible to the surveillance camera,'' recounted a still chuckling Cmdr. Ronald Freeman of the investigations branch.
This may be betraying my ignorance, but I thought that the basic security model behind android held that one app couldn't see another app's code or private data. The sdcard is general storage, so all apps with sdcard permissions can see everything on the card, but mostly what is stored on the sdcard is not security critical anyway. Another caveat is that if you've rooted your phone then you're adult enough to look after yourself.
So, how is a virus scanner supposed to work? It will never be able to see any of the other apps. Similarly with a firewall - it wont be able to see when another app connects to the net. The only way it could work would be to break out of the standard security systems.
This suggests to me that most of these 'security apps' are scams (not counting stuff for rooted phones -- but rooting is itself a security risk. e.g. With most of the rooting mechanisms, if you hook a usb cable to a rooted phone then you get a root shell without any password protection - mmmmm security).
The 'security apps' that I could see working are:
- Apps that monitor the SIM ID so that they know when another SIM is inserted.
- Apps that allow remote phone tracking.
It is possible to detect some changes in phone state and quickly try and correct things, or at least alert the user to the issue. This is how the apps that add password screens/PINs to the front of other apps work - they detect the other app being brought to the front and quickly overlay their pin screen over the top. But this seems weak for general protection.
So, what specific protection do people think they're getting from these security programs? Which of the (non-root) programs more than a placebo, and how do they work?
I'm also on internode and use a NetComm NB6Plus4 to get my IPv6 over ADSL. The NB6Plus4 needs a firmware update that internode have on their web site. You then need to use a slightly modified PPPoE login to tell internode you want IPv6.
Note that home ADSL ipv6 is currently a trial and unsupported. But it works well:).
The real holdup with home equipment seems to be getting home modem/routers that support IPv6 (routing and PPPoE). I would have thought this would be fixable with firmware updates though. (Or you can stick your modem in bridge mode and use a PC with PPPoE IPv6 support - but that's unlikely to be a popular option.)
Congrats to the teams that did well. I know a bunch of Australian teams that looked into entering and decided not to because:
a) It was an engineering challenge more than a research challenge,
b) It was closer to that ethical line of making killer robots than, say, the DARPA Grand Challenge autonomous vehicle competition,
c) There was an extremely compressed timeline to actually make anything, and
d) The prize is mostly prestige. i.e. It wouldn't come anywhere near the development costs even for the teams that won.
So, it was a less than perfect competition. But that also means that the teams that did well in it did well under difficult conditions, so good for them.:)
- This guy was a state, not a federal, politician. This is important for two reasons: i) it is the federal govt that is looking to introduce the filter, not the state govt, and ii) The federal govt is 1 by-election away from a change in majority, so that would be much more tricky. iii) The state govt is so dysfunctional that noone really cares any more - they're going to lose the next election and this guy might well be happy to go now before the rush.
- This story is a little old. Although only a couple of weeks old, so for Slashdot maybe that is ok.
- There were a number of people 'caught' by the audit, including staffers for one of the more religious parties (no surprise there). But after that it came out that the audit itself may have had some problems...
It is legislation, not case law. Section 35(6) of the Australian Copyright Act 1968: "Where a literary... work... is made by the author in pursuance of the terms of his or her employment by another person under a contract of service or apprenticeship, that other person is the owner of any copyright subsisting in the work...." Computer Software is a 'literary work' for the purposes of the act.
Note that it isn't all code written by an employee, just code written for your job.
Yup. And this is why we don't have robots that move about the world yet. The world is messy and simple rules applied with no ambiguity don't work well.
Oh for Mod points. Arctanx is right. There are at least three quite separate issues here:
a) Stephen Conroy's policies are woefully misguided. They will have the opposite effect on child protection to that publicly claimed, causing an increase the pain and suffering of many children.
b) Even so, registering this protest site is not a good response. Have the site, but put it in stephenconroysucks.com.au.
- I much prefer Australia's rather stricter DNS rules to the loose ones in the US. This is not a freedom of speech issue, it's a truth in advertising issue.
- Applying for a business name as a workaround is not a good thing. It will make the laws either a) more complex, or b) more vague so that they cover all the misleading behaviour.
c) Regardless of whether the AuDA rules are good, the AuDA implementation looks like it needs some work. Having said that, I'm a fan of stricter policing of au DNS rules. Noone was taking the server down - they were simply saying it can't have a particular domain name.
The TSA causes so much hassle in the US that it is a drain on the country. Some countries, like Australia, have recently started loosening their restrictions (I hear that metal cutlery is back).
Imagine you're a terrorist organisation. You find some idiot willing to do this. If they succeed, great. If they fail, then you're still going to inconvenience the entire travelling world in the middle of one of the busiest times of the year.
Look at the lists of new 'safety rules' that have been instituted. For Al-Qaeda, that's not a bad consolation prise.
If I recall correctly, MS at one point tried to say that, if something like this happened, you'd have to release all your source code. Now we find that MS knows that you only have to release the source code of the program in question. Big difference.
I don't believe that's correct. You need to stop infringing the copyright - that means either obey the terms of the license or stop distributing (and deal with the consequences of the limited distribution you already made).
I hope this doesn't help the bogus 'GPL is dangerous, an outside contractor can make you reveal your code' meme to spread.
I always heard that the imperial system was 'intuitive' because it allowed lots of ways of packing things. If you have 12 items, then you can make a 4 x 3 box of them and it is almost square. You could also do 2x6 (they pack eggs that way).
You could split the between imperial and metric and switch to a consistent base 12. You'd get all the nice features of a consistent system, and you'd still have nice packing for the supermarket. "Honey, I broke an egg. Now there are only B left."
I would recommend base 16, but noone ever uses that.
Online news has been stuck in a prisoner's dilemma situation (from their POV). If everyone charged for news, then they'd be OK. When only some people charge for news, those that charge lose their audience. That drives the system to the equilibrium of noone charging for news. From the consumer's POV this is a good thing.
Because Murdoch owns so much of the news, he might be able to break out of the current poor (for newspaper publishers) equilibrium. Of course, if he can do so then he's pretty much demonstrated that he has enough of a monopoly that market power isn't working. There would be evidence for an anti-trust case against him.
The other problem with all this is that it assumes that the problem newspapers are having with revenue is caused by the cannibalisation of the print editions by the online editions. I understand, although I cannot provide evidence, that the real problem is that the classified market has gone away. The newspapers lunch got eaten by eBay and Craigslist, not cannibalised by their own online offerings. And if this is true, then raising prices for consumers might increase revenue, but it wont return it to where it was.
TiniWiki also does this. I haven't done a detailed comparison with the one in the article, but I last time I looked TinyWiki was pretty good. They had two advantages over some other similar products: a) they had more of Wikipedia, not just a cut-down or old selection, and b) they could do incremental updates.
Many safari plugins use InputManager hacks to patch safari. That can be dangerous, and I don't have the time to really check which ones are safe and which ones aren't.
In contrast, I believe that ClickToFlash is just another web plugin (the same as flash itself). It is used instead of the normal flash plugin. It displays a plain grey box. When clicked on, that grey box loads the normal flash plugin. The result is a flash blocker that works in safari using standard APIs.
Just to clarify, MarkvW said the "Parent article" is a full of gross misstatements. I assume MarkvW is referring to Bill Snyder's "Parent article" (the first link in the summary and referencing Lawyer Jonathan Moskin's coments), not rewt66's "Parent post" of his knowledge of the case (which matches my memory of the case).
I don't think Snyder's article is full of "GROSS misstatements". I think it is full of the sort of ordinary misstatements you get when a well meaning journalist is reporting on something they don't really get. It is full of false balance, and it leaves out some important details stating things like "The facts in the case are somewhat muddled". I suspect Snyder's understanding is more muddled than the facts, but I think he did an OK (but not good) job.
It looks vaguely android-esque. The clock looks the same, but there isn't a little tabby thing on the bottom of the screen - except there is a little tabby thing on one side.
Is it Android? Or is cloned from the Android UI? Or is it just parallel evolution?
Samsung has announced using Android for other phones.
It seems to me that part of the issue here is that you're trying to form a single ranking of all the papers/journals, and there might not be one. Netflix doesn't try to form a single ranking of all movies, they try to find the ones that a particular individual will like - a personalised definition of good.
This allows the crackpots to have their own definition of 'good', and there is nothing wrong with that.
For individual researchers this approach would probably work very well. Funding bodies would need to specify more constraints than just that they want "good research" to get a useful answer. Figuring out what those extra constraints 'should' be is an interesting question.
The Wolfram page notes two approaches. One is a heuristic approach that works well until the board gets large (76x76 is the size quoted). This is the algorithm implemented by Python code.
The second approach is only referenced on the Wolfram page, not described in any detail. The text on that algorithm from the Wolfram page is:
Recently, Conrad et al. (1994) discovered another linear time algorithm and proved that it solves the Hamiltonian path problem for all n >= 5. The Conrad et al. algorithm works by decomposition of the chessboard into smaller chessboards (not necessarily square) for which explicit solutions are known. This algorithm is rather complicated because it has to deal with many special cases, but has been implemented in Mathematica by A. Roth.
Sounds hard to implement in 60 lines of python - especially when most of the 60 lines is UI.
Slashdot Mirror
Govt: "Give me all your URLs"
ISPs: "No"
Govt: "OK, will you give me half?"
ISPs: "OK"
This is not a win. It is true that the government is no longer proposing new legislation. This is because they seem to be getting much of what they want without new legislation. In particular, it appears there will be a new industry "Code of Practice" which will then have force under current legislation (think of it as like changing a regulation). There is also going to be a 'police request' to the ISPs. It isn't clear to me at this point which of these, or how the combination, leads to enforceability, but it seems all Aus ISPs are now going to be filtered.
Good:
- The black-list will be the Interpol black-list, which I understand is at least reasonably well monitored and is quite small. It is also outside of political control in Australia.
- I'm guessing the filtering will be DNS based for domains on the list, forwarding them to a proxy that checks the URL. This means it will not affect people not accessing those domains.
Bad:
- We have a filter. This will have two effects: it will lead to worse enforcement of child abuse as people get complacent. It will cause the occasional problem like the Wikipedia/Scorpians cover (see http://en.wikipedia.org/wiki/Internet_Watch_Foundation_and_Wikipedia )
This is a shame. Child porn sites stay up longer than phishing sites because banks actually want the phishing sites down, not just covered up.
It will also be interesting to see what effect DNS-SEC has on this, but that is in the longer term.
Give each country its own DNS. [snip]
Each country already has its own DNS. Country code domain names have been around for a long time. Maintaining the root servers that point to the country codes doesn't need to be an automated system. I'd hand it over to the group that agrees on the country codes: the UN.
What you're really suggesting is getting rid of non-country code domain names. All those 'blah.com' addresses would need to choose one or more country codes to occupy... 'blah.com.us' or 'blah.co.uk' or ... . This would be an improvement on the current situation, but it would also be a huge upheaval and I'm not sure the long term benefits are worth the short term cost.
But even once you've done that you still haven't solved all the problems. If a UK resident purchases from a US shop, whose laws apply? US, UK or international law? It is at least clear to customer that they're purchasing from an overseas entity, but that doesn't solve the legal questions. Things could get messier still: what happens if a .co.uk domain name points to a server in the US and there is an Australian customer? Whose laws apply then?
And lets not get started on the allocation of IP addresses... :)
For every complex problem there is a solution that is simple, neat and wrong.
Reminds me of McArthur Wheeler: (from http://plus.sites.post-gazette.com/index.php/component/content/article/62/103182-pittsburgher-stupidity-in-the-news-the-mcarthur-wheeler-effect- )
At 5 feet 6 inches and about 270 pounds, McArthur Wheeler is an easily recognizable man — even when wearing lemon juice on his face.
That certainly came as a surprise to Wheeler, 45, of Versailles Street, McKeesport. He was incredulous in April when Pittsburgh robbery detectives told him that he had been identified in surveillance photographs as one of the two men who robbed two banks in Brighton Heights and Swissvale on Jan. 6.
"But I wore the lemon juice. I wore the lemon juice,'' a puzzled Wheeler told the even more puzzled detectives.
The detectives' confusion turned to incredulity as Wheeler explained about his would-be lemon aid.
"Someone told him that if you put lemon juice on your face it makes you invisible to the surveillance camera,'' recounted a still chuckling Cmdr. Ronald Freeman of the investigations branch.
This may be betraying my ignorance, but I thought that the basic security model behind android held that one app couldn't see another app's code or private data. The sdcard is general storage, so all apps with sdcard permissions can see everything on the card, but mostly what is stored on the sdcard is not security critical anyway. Another caveat is that if you've rooted your phone then you're adult enough to look after yourself.
So, how is a virus scanner supposed to work? It will never be able to see any of the other apps. Similarly with a firewall - it wont be able to see when another app connects to the net. The only way it could work would be to break out of the standard security systems.
This suggests to me that most of these 'security apps' are scams (not counting stuff for rooted phones -- but rooting is itself a security risk. e.g. With most of the rooting mechanisms, if you hook a usb cable to a rooted phone then you get a root shell without any password protection - mmmmm security).
The 'security apps' that I could see working are:
- Apps that monitor the SIM ID so that they know when another SIM is inserted.
- Apps that allow remote phone tracking.
It is possible to detect some changes in phone state and quickly try and correct things, or at least alert the user to the issue. This is how the apps that add password screens/PINs to the front of other apps work - they detect the other app being brought to the front and quickly overlay their pin screen over the top. But this seems weak for general protection.
So, what specific protection do people think they're getting from these security programs? Which of the (non-root) programs more than a placebo, and how do they work?
I'm also on internode and use a NetComm NB6Plus4 to get my IPv6 over ADSL. The NB6Plus4 needs a firmware update that internode have on their web site. You then need to use a slightly modified PPPoE login to tell internode you want IPv6.
See http://ipv6.internode.on.net/configuration/adsl-faq-guide/
Note that home ADSL ipv6 is currently a trial and unsupported. But it works well :).
The real holdup with home equipment seems to be getting home modem/routers that support IPv6 (routing and PPPoE). I would have thought this would be fixable with firmware updates though. (Or you can stick your modem in bridge mode and use a PC with PPPoE IPv6 support - but that's unlikely to be a popular option.)
Congrats to the teams that did well. I know a bunch of Australian teams that looked into entering and decided not to because:
a) It was an engineering challenge more than a research challenge,
b) It was closer to that ethical line of making killer robots than, say, the DARPA Grand Challenge autonomous vehicle competition,
c) There was an extremely compressed timeline to actually make anything, and
d) The prize is mostly prestige. i.e. It wouldn't come anywhere near the development costs even for the teams that won.
So, it was a less than perfect competition. But that also means that the teams that did well in it did well under difficult conditions, so good for them. :)
And if you go there and click on the next link it tells you to use Software Update to get Apple's JVM.
A few things:
- This guy was a state, not a federal, politician. This is important for two reasons: i) it is the federal govt that is looking to introduce the filter, not the state govt, and ii) The federal govt is 1 by-election away from a change in majority, so that would be much more tricky. iii) The state govt is so dysfunctional that noone really cares any more - they're going to lose the next election and this guy might well be happy to go now before the rush.
- This story is a little old. Although only a couple of weeks old, so for Slashdot maybe that is ok.
- There were a number of people 'caught' by the audit, including staffers for one of the more religious parties (no surprise there). But after that it came out that the audit itself may have had some problems...
I highly recommend this link on why transparency is not enough.
It is legislation, not case law. Section 35(6) of the Australian Copyright Act 1968: "Where a literary ... work ... is made by the author in pursuance of the terms of his or her employment by another person under a contract of service or apprenticeship, that other person is the owner of any copyright subsisting in the work ... ." Computer Software is a 'literary work' for the purposes of the act.
Note that it isn't all code written by an employee, just code written for your job.
IANAL, but you can look at the law yourself: http://www.austlii.edu.au/au/legis/cth/consol_act/ca1968133/s35.html
Here's a little more. http://www.youtube.com/watch?v=ihoWNnEZ5zg Again, it isn't a full doco, but it shows what the place looks like.
The video isn't playing for me. Is there another version out there? I found this: http://www.youtube.com/watch?v=EMCtzuEoOlM , but it's only a short ad.
Yup. And this is why we don't have robots that move about the world yet. The world is messy and simple rules applied with no ambiguity don't work well.
Oh for Mod points. Arctanx is right. There are at least three quite separate issues here:
a) Stephen Conroy's policies are woefully misguided. They will have the opposite effect on child protection to that publicly claimed, causing an increase the pain and suffering of many children.
b) Even so, registering this protest site is not a good response. Have the site, but put it in stephenconroysucks.com.au.
- I much prefer Australia's rather stricter DNS rules to the loose ones in the US. This is not a freedom of speech issue, it's a truth in advertising issue.
- Applying for a business name as a workaround is not a good thing. It will make the laws either a) more complex, or b) more vague so that they cover all the misleading behaviour.
c) Regardless of whether the AuDA rules are good, the AuDA implementation looks like it needs some work. Having said that, I'm a fan of stricter policing of au DNS rules. Noone was taking the server down - they were simply saying it can't have a particular domain name.
The TSA causes so much hassle in the US that it is a drain on the country. Some countries, like Australia, have recently started loosening their restrictions (I hear that metal cutlery is back).
Imagine you're a terrorist organisation. You find some idiot willing to do this. If they succeed, great. If they fail, then you're still going to inconvenience the entire travelling world in the middle of one of the busiest times of the year.
Look at the lists of new 'safety rules' that have been instituted. For Al-Qaeda, that's not a bad consolation prise.
If I recall correctly, MS at one point tried to say that, if something like this happened, you'd have to release all your source code. Now we find that MS knows that you only have to release the source code of the program in question. Big difference.
I don't believe that's correct. You need to stop infringing the copyright - that means either obey the terms of the license or stop distributing (and deal with the consequences of the limited distribution you already made).
I hope this doesn't help the bogus 'GPL is dangerous, an outside contractor can make you reveal your code' meme to spread.
I always heard that the imperial system was 'intuitive' because it allowed lots of ways of packing things. If you have 12 items, then you can make a 4 x 3 box of them and it is almost square. You could also do 2x6 (they pack eggs that way).
You could split the between imperial and metric and switch to a consistent base 12. You'd get all the nice features of a consistent system, and you'd still have nice packing for the supermarket. "Honey, I broke an egg. Now there are only B left."
I would recommend base 16, but noone ever uses that.
Or someone could prove it doesn't happen in their kernel...
http://tech.slashdot.org/story/09/08/13/0827231/Worlds-First-Formally-Proven-OS-Kernel
Online news has been stuck in a prisoner's dilemma situation (from their POV). If everyone charged for news, then they'd be OK. When only some people charge for news, those that charge lose their audience. That drives the system to the equilibrium of noone charging for news. From the consumer's POV this is a good thing.
Because Murdoch owns so much of the news, he might be able to break out of the current poor (for newspaper publishers) equilibrium. Of course, if he can do so then he's pretty much demonstrated that he has enough of a monopoly that market power isn't working. There would be evidence for an anti-trust case against him.
The other problem with all this is that it assumes that the problem newspapers are having with revenue is caused by the cannibalisation of the print editions by the online editions. I understand, although I cannot provide evidence, that the real problem is that the classified market has gone away. The newspapers lunch got eaten by eBay and Craigslist, not cannibalised by their own online offerings. And if this is true, then raising prices for consumers might increase revenue, but it wont return it to where it was.
TiniWiki also does this. I haven't done a detailed comparison with the one in the article, but I last time I looked TinyWiki was pretty good. They had two advantages over some other similar products: a) they had more of Wikipedia, not just a cut-down or old selection, and b) they could do incremental updates.
I prefer ClickToFlash to many of the other flash blockers for safari:
http://github.com/rentzsch/clicktoflash/tree/master
Many safari plugins use InputManager hacks to patch safari. That can be dangerous, and I don't have the time to really check which ones are safe and which ones aren't.
In contrast, I believe that ClickToFlash is just another web plugin (the same as flash itself). It is used instead of the normal flash plugin. It displays a plain grey box. When clicked on, that grey box loads the normal flash plugin. The result is a flash blocker that works in safari using standard APIs.
Just to clarify, MarkvW said the "Parent article" is a full of gross misstatements. I assume MarkvW is referring to Bill Snyder's "Parent article" (the first link in the summary and referencing Lawyer Jonathan Moskin's coments), not rewt66's "Parent post" of his knowledge of the case (which matches my memory of the case).
I don't think Snyder's article is full of "GROSS misstatements". I think it is full of the sort of ordinary misstatements you get when a well meaning journalist is reporting on something they don't really get. It is full of false balance, and it leaves out some important details stating things like "The facts in the case are somewhat muddled". I suspect Snyder's understanding is more muddled than the facts, but I think he did an OK (but not good) job.
It looks vaguely android-esque. The clock looks the same, but there isn't a little tabby thing on the bottom of the screen - except there is a little tabby thing on one side.
Is it Android? Or is cloned from the Android UI? Or is it just parallel evolution?
Samsung has announced using Android for other phones.
It seems to me that part of the issue here is that you're trying to form a single ranking of all the papers/journals, and there might not be one. Netflix doesn't try to form a single ranking of all movies, they try to find the ones that a particular individual will like - a personalised definition of good.
This allows the crackpots to have their own definition of 'good', and there is nothing wrong with that.
For individual researchers this approach would probably work very well. Funding bodies would need to specify more constraints than just that they want "good research" to get a useful answer. Figuring out what those extra constraints 'should' be is an interesting question.
The Wolfram page notes two approaches. One is a heuristic approach that works well until the board gets large (76x76 is the size quoted). This is the algorithm implemented by Python code.
The second approach is only referenced on the Wolfram page, not described in any detail. The text on that algorithm from the Wolfram page is:
Sounds hard to implement in 60 lines of python - especially when most of the 60 lines is UI.