Solution From Rory Hart Of ExitReality To Protect Your Works

As you should know, Hiperia3D News discovered a security flaw in ExitReality. Their search engine has already indexed the worlds and models of many VRML worldbuilders, without asking their previous permission. Read this here.

But Rory Hart, the Head of Development of ExitReality, has given an answer today in the comments of my article. This note has also be sent to some virtual worlds forums. They provide you two ways to prevent your objects and models to be incorporated to their search engine: the robots.txt file and the referrer in the http header.

This is a good starting point to correct something that has upset many worldbuilders, and I congratulate Rory Hart for this act.

Here's his quote, and after it, my comments and feedback about this, as requested.

"We use a crawler to collect urls of wrl files for linking in our web 3d search engine. Its useragent string is ExitRealitySearchAgent.

This crawler obeys the normal robot.txt set of rules for search engines as described in:

To address the issue that has been describe on this forum I suggest these entries to tell the ExitReality spider not to index vrml worlds on your domain:

User-agent: ExitRealitySearchAgent
Disallow: /*.wrl$

User-agent: ExitRealitySearchAgent
Disallow: /*.wrz$

To ensure your site is crawled quickly I would also suggest submitting your domain for a priority spidering at the following url once your robot.txt file has been updated.

Also I noted Paul Aslin's post on the public web 3d mailing lists in regards to vrml browsers providing referrer in the http header. We already provide a referrer in the http headers when our browser makes a request. This contains the referring file (whether it is html or vrml) and was implemented to help prevent deep linking.

Your webserver can be configured to disallow deeplinking by checking this referrer and constraining it to referrers that you allow. Configuration of this is dependant on your hosting service and the type of http server they use.

All this information will be fasttracked to our knowledgebase the assist others who share these concerns.

We have read your comments on our web 3d search and are discussing your suggestions. I would like to take this opportunity to state our intentions for the web 3d search. We do not intend it to pass off others work as our own, rather the search is meant to provide an accessible way for people to find and view 3d content.

Thanks for your interest and please feel free contribute feedback."

-Rory Hart, Head of Development of ExitReality, in the comments of Hiperia3D News


ExitReality should not index anything without the prior consent of the builder. The builder should not do anything to prevent this.

Other software companies, like Vivaty or 3DXplorer, have their own object catalogs built by themselves, or rewarding the creators (that have previously given permission to use their objects and scenes).
ExitReality takes a great advantage over these other companies taking objects for their software without asking.

This is not how copyright works. You should ask first, and then use -- not the opposite: use first unless I forbid you to do it.

Anyway, Hiperia3D News' purpose is just give the news, not to take part on anything of this. Now it's time for worldbuilders to negotiate this. ExitReality has taken a first step and one thing is for sure: they listen the feedback.

-Jordi R. Cardona-

© 2008 by Jordi R. Cardona. Link to this post without copying the text.

If you liked this post, get updates of Hiperia3D News for FREE


S(z)arah said...

I enjoy your point of view and mine meet ... and the common sense and the law too :)
I think it's time to publish all that for my own french readers : tomorrow you'll receive a fine backlink.

Taranis said...

Sad news I have to tell : according to the raw logs from my servers , it appears that the exitreality spider does just not exists : it's a mozilla that scan url and sites submitted to their site.

It does NOT look at the file robots.txt.

According to this, I can say that beside being deeplinkers and content misusers, exitreality are liars.

Jordi R Cardona said...

Hi Taranis,

That's strange. How do you distinguish that the robot doesn't exist? Have others confirmed this?

Taranis said...

Yes it has to be confirmed, this will be done soon.
How do I see the robot does not exists ? I check my website logs and see what type of robots has been checking the robots.txt file.
According to ER, I should have seen a ERsearchagent looking to my robots.txt file : I saw NONE.
At the same time, a curious IP from Australia is looking at my site files ( not to mention, it does not look to the robots.txt file ) coming from revemonde forum ... The same IP checked also all my wrl files from my site on revemonde.

So lets' just wait a bit to get a confirmation. It is clear anyway that I did not see any spider called ERsearchagent looking to my robots.txt at the time I put it and before ER de-indexed my files.

Anonymous said...

FYI- google loses in court- again.

Jordi R Cardona said...

Yes, but I think that's a bit bad, I would never had demanded Google, because Google links to your true domain name and page, and says:"Image may be scaled down and subject to copyright."

If someone that is too stupid uses one of those images, you can request their host to delete their domain or page.

Many sites get visitors to their site from Google Image Search, visitors that see who is the owner, etc.

It's NOT the same as ExitReality. The comparison is totally off-topic.

Rory Hart is giving a good priority on this, providing advancement in the solution of this issue, as I'll post in a future article.

It would have been better that they ask before use, but if they provide a good solution and we spread it enough, it would be a good ending.

Jordi R Cardona said...

This may be a good moment to beta test this:

and I insist: ER should just show the copyright info displayed on WorldInfo in many old vrml files.

But just go and test their methods, and tell us what you get.

Post a Comment