Website Statistics

[Vetran]

Re: re

I can do cost effective - not cheap

[reynolds]

re

Now with image ripping:

www.mycgiserver.com/~auto...bscan.html

[fiddleabout]

Re: re

Sadly it doesn't work through the proxy here.

I guess you just serve the img tag and don't pull the image and re-serve or it could be very useful for checking some of the educational sites the rabbit recommends which are blocked by webwasher.

Now there's a freebie business idea for you

[reynolds]

re

It should work through all browsers & proxys??

How would you go about content blocking? I mean you can scan images but how would you actually get the program to tell if they are unsuitable or not?

[fiddleabout]

Re: re

Maybe it should but it doesn't - I notice that the results page specifies port 9090 - probably all ports other than standard ones are blocked.

I think you misunderstand my comment on graphics. The site I work at uses an additional proxy called WebWasher (also DynaBlocker or something very like that) which prevents access to a humanly selected set of urls.

So - if you actually downloaded the image and then re-served it from your site it would bypass the censor It used to be possible to check old copies of "educational" sites using www.archive.org but they got wise to that and blocked it too. If you produced a service no doubt that would be blocked too eventually so it was more of a joke than a request.

[reynolds]

re

I've ported the code to an application now, and it now follows links so depending where I set root it will literally return gigabytes of data. The threads are a little out of control at the moment so I need to regulate them in a ThreadGroup. Not quite sure whether this could have business value, I guess it's just a matter of how I use the information gathered and whether I point it to certain domains and markets. Google tends to not specialise in areas but sweep up everything so there could be a market in domain-specific data-mining.