Specify XML application to allow of classification any page. Of
course, user must add classification institution to white list.
We can create two ways to work of this mechanism:
1) White list
2) Black list
Institution, who classify pages can use one or both. Black list
requires to add new button into web browsers, called "report bad
page".
Copy of White/Black list can be shipped onto Mozilla's servers, but
(of course) only for few classification institution. So in some cases
if we trust classification institution, no information onto it servers
will be sends.
To realize this idea we requires to create new tag:
<classify role='subject' type='childhood' url="http://insytution.org"
approval_date="date" description="Our page avoids violence, etc.">
<classify role='subject' type='adults only' url="http://
insytution.org" approval_date="date">
Description will be optional. It would prevent blocking good pages by
some filters. If page are classified by trusted organization, we won't
block it.
Of course I'm scared too, because it brinks new way of control, what
people does on net, but private computers shouldn't been touched.
Another way to use it is only watching on type. If page is classified
as contains violence, web browser would told: "you cannot even seen
this page"(of course it must working in kiosk mode).
Can you explain the specific use case for this extension to the browser DOM a bit more? Is this about blocking certain pages? If so - how do you think this would be easier to implement than the usual blacklist services?
Lachu wrote:
> Specify XML application to allow of classification any page. Of
> course, user must add classification institution to white list.
> We can create two ways to work of this mechanism:
> 1) White list
> 2) Black list
> Institution, who classify pages can use one or both. Black list
> requires to add new button into web browsers, called "report bad
> page".
> Copy of White/Black list can be shipped onto Mozilla's servers, but
> (of course) only for few classification institution. So in some cases
> if we trust classification institution, no information onto it servers
> will be sends.
> To realize this idea we requires to create new tag:
> <classify role='subject' type='childhood' url="http://insytution.org"
> approval_date="date" description="Our page avoids violence, etc.">
> <classify role='subject' type='adults only' url="http://
> insytution.org" approval_date="date">
> Description will be optional. It would prevent blocking good pages by
> some filters. If page are classified by trusted organization, we won't
> block it.
> Of course I'm scared too, because it brinks new way of control, what
> people does on net, but private computers shouldn't been touched.
> Another way to use it is only watching on type. If page is classified
> as contains violence, web browser would told: "you cannot even seen
> this page"(of course it must working in kiosk mode).
-- Pascal Finette | Catalyst | Mozilla Labs | +1.650.903.0800x656
It will been used in many cases. Google and any other robots will use
this. Web based news clients can use this. Web browsers can use this.
Simple use case:
Patric have used Firefox and page filtering program. He are 15 year
old. Page filtering(looking for images) program will told: adults
only, but page are classify by trusted foundation to green page.
Better use case:
The terms of service told, it bases on U.S.A law. Imagine that it is
auction service. Additionally if administrator or term of service
break/hangs U.S.A law, it doesn't be properly classified.
Another use case:
Each forum can be classified as forum if there no bad/violence
language.
Other use case:
Big organization can bought insurance of web page. If web page don't
contains helper for disabled people, we know, where we can go.
On Nov 2, 7:23 pm, Pascal Finette <pfine...@mozilla.com> wrote:
> Can you explain the specific use case for this extension to the browser
> DOM a bit more? Is this about blocking certain pages? If so - how do
> you think this would be easier to implement than the usual blacklist
> services?
> Warm regards
> P'
> Lachu wrote:
> > Specify XML application to allow of classification any page. Of
> > course, user must add classification institution to white list.
> > We can create two ways to work of this mechanism:
> > 1) White list
> > 2) Black list
> > Institution, who classify pages can use one or both. Black list
> > requires to add new button into web browsers, called "report bad
> > page".
> > Copy of White/Black list can be shipped onto Mozilla's servers, but
> > (of course) only for few classification institution. So in some cases
> > if we trust classification institution, no information onto it servers
> > will be sends.
> > To realize this idea we requires to create new tag:
> > <classify role='subject' type='childhood' url="http://insytution.org"
> > approval_date="date" description="Our page avoids violence, etc.">
> > <classify role='subject' type='adults only' url="http://
> > insytution.org" approval_date="date">
> > Description will be optional. It would prevent blocking good pages by
> > some filters. If page are classified by trusted organization, we won't
> > block it.
> > Of course I'm scared too, because it brinks new way of control, what
> > people does on net, but private computers shouldn't been touched.
> > Another way to use it is only watching on type. If page is classified
> > as contains violence, web browser would told: "you cannot even seen
> > this page"(of course it must working in kiosk mode).
Another use case:
Some organization working in the cloud. It can put this tag on each
own web page and use white list of url's and IP addresses, but (of
course) better using fingerprints.
On Nov 2, 7:23 pm, Pascal Finette <pfine...@mozilla.com> wrote:
> Can you explain the specific use case for this extension to the browser
> DOM a bit more? Is this about blocking certain pages? If so - how do
> you think this would be easier to implement than the usual blacklist
> services?
> Warm regards
> P'
> Lachu wrote:
> > Specify XML application to allow of classification any page. Of
> > course, user must add classification institution to white list.
> > We can create two ways to work of this mechanism:
> > 1) White list
> > 2) Black list
> > Institution, who classify pages can use one or both. Black list
> > requires to add new button into web browsers, called "report bad
> > page".
> > Copy of White/Black list can be shipped onto Mozilla's servers, but
> > (of course) only for few classification institution. So in some cases
> > if we trust classification institution, no information onto it servers
> > will be sends.
> > To realize this idea we requires to create new tag:
> > <classify role='subject' type='childhood' url="http://insytution.org"
> > approval_date="date" description="Our page avoids violence, etc.">
> > <classify role='subject' type='adults only' url="http://
> > insytution.org" approval_date="date">
> > Description will be optional. It would prevent blocking good pages by
> > some filters. If page are classified by trusted organization, we won't
> > block it.
> > Of course I'm scared too, because it brinks new way of control, what
> > people does on net, but private computers shouldn't been touched.
> > Another way to use it is only watching on type. If page is classified
> > as contains violence, web browser would told: "you cannot even seen
> > this page"(of course it must working in kiosk mode).
|
