Just a quick message to say that all ideas for this project are very welcome, and that as coders we appreciate communication with anyone interested in any part of the project!
Regards,
William Cheung
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Greetings,
Saw your want ad for an IE programmer. I've got quite a bit of experience hacking IE, but I wanted to ask a couple of things about the project....
First off - what kind of filtering/safety features are you wanting to offer? IE already supports ICRA - see http://www.icra.org/ . What more are you wanting to offer? In other words - what can this project offer the public that isn't there already?
Next - why Internet Explorer? If you were targetting only Windows, it'd make some sense for development speed and market share - but you're also targetting Macs according to the project settings.
Microsoft is *very* reluctant to release integration information or tools for IE on Macintosh in my experience - I've wasted a lot of time in meetings with them trying to get them to in the past. Basically, unless you have a customer base the size of AOL's, don't expect to be able to embed and modify the behavior of IE on a Macintosh like you can with IWebBrowser2 on Windows.
My initial thought is that adding ICRA support and lock-down controls to Mozilla would be the best route to providing a free, open source, multiplatform solution for a child-safe browser. However, I haven't used the child-protection type software myself really, so I may be missing something.
Cheers,
Mike Ellison
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
The ad was for an IE experience person because I thought it a good place to start. I would like this to be a cross platform OS software package which will work on Windows, Mac and Linux.
I have looked at the ICRA info and I found that the idea of a hand rated system which tags websites to be interesting but restrictive. The idea of Project Innocence is to allow the people the right to control what they or their children are exposed to. Which means that the filter we are trying to build will work on any site based on the content it has and that if a database of websites is kept for allowable and banned sites then the end users will do the rating.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Thanks for the ICSA info - I haven't much information with regards to how the IE content rating system works, but I shall peruse! :)
Just to reinforce Mark's idea - I think that the idea of a generic contents determination system - as opposed to a manual, heuristic database - may bring certain advantages in that some pages may be paraphrased, etc or have their addresses modified which may be performed in order to fool a database of pages.
I'm afraid my above points may sound a bit too simplisitic and I agree with the point about Microsoft's reluctance to release code information (and rightly so). Perhaps more of the project emphasise is server-side, and classifies through machine learning?
Best wishes,
William Cheung
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi all,
Just a quick message to say that all ideas for this project are very welcome, and that as coders we appreciate communication with anyone interested in any part of the project!
Regards,
William Cheung
Greetings,
Saw your want ad for an IE programmer. I've got quite a bit of experience hacking IE, but I wanted to ask a couple of things about the project....
First off - what kind of filtering/safety features are you wanting to offer? IE already supports ICRA - see http://www.icra.org/ . What more are you wanting to offer? In other words - what can this project offer the public that isn't there already?
Next - why Internet Explorer? If you were targetting only Windows, it'd make some sense for development speed and market share - but you're also targetting Macs according to the project settings.
Microsoft is *very* reluctant to release integration information or tools for IE on Macintosh in my experience - I've wasted a lot of time in meetings with them trying to get them to in the past. Basically, unless you have a customer base the size of AOL's, don't expect to be able to embed and modify the behavior of IE on a Macintosh like you can with IWebBrowser2 on Windows.
My initial thought is that adding ICRA support and lock-down controls to Mozilla would be the best route to providing a free, open source, multiplatform solution for a child-safe browser. However, I haven't used the child-protection type software myself really, so I may be missing something.
Cheers,
Mike Ellison
The ad was for an IE experience person because I thought it a good place to start. I would like this to be a cross platform OS software package which will work on Windows, Mac and Linux.
I have looked at the ICRA info and I found that the idea of a hand rated system which tags websites to be interesting but restrictive. The idea of Project Innocence is to allow the people the right to control what they or their children are exposed to. Which means that the filter we are trying to build will work on any site based on the content it has and that if a database of websites is kept for allowable and banned sites then the end users will do the rating.
Thanks for the ICSA info - I haven't much information with regards to how the IE content rating system works, but I shall peruse! :)
Just to reinforce Mark's idea - I think that the idea of a generic contents determination system - as opposed to a manual, heuristic database - may bring certain advantages in that some pages may be paraphrased, etc or have their addresses modified which may be performed in order to fool a database of pages.
I'm afraid my above points may sound a bit too simplisitic and I agree with the point about Microsoft's reluctance to release code information (and rightly so). Perhaps more of the project emphasise is server-side, and classifies through machine learning?
Best wishes,
William Cheung