Jump to content

Loopy Idea


Recommended Posts

poofsplz.gif?1

 

-ignore, I had posted a link, then doubted its usefulness and removed it again.

Peter's post somehow led me to wonder what would happen if I created a post, then edited it to insert a link to itself. Shouldn't be anything strange, but considering some other things I've observed while immersed in this Invision software, I just have to try it. :)

 

OK, here goes. What happens if I follow this link? :unsure:

 

(Now I try it...)

 

Interesting! When I open the thread, it creates a new level in my browser, as I think it should. Then when I click on the link, it creates yet another level in my browser, but further clicks on the link don't add further levels. But I guess it's my browser rather than Invision that's responsible.

 

----

 

Oh, wait a minute! The first time, the URL is just for the thread, while the second time it's for the specific post. They just happen to give the same view when there's only this one post in the thread. So, a variation: link to the thread rather than to the specific post. Here.

 

Yep, that's it. No extra level when I click on the "Here" link from the full thread. That makes sense. The browser doesn't repeat a duplicate, but of course only if it's an exact duplicate. I guess I should have expected that.

 

Well, now I'd better get back to doing something useful. B)

As for those of you who bothered to read this post, we're both to blame, but if you take the trouble to duplicate my experiment, I will not accept responsibility for your foolishness... only for my own. :ph34r:

 

Edited by JimLucas
Link to comment
Share on other sites

I once did something extremely antisocial as a proof of concept in an obscure corner of Usenet. Usenet messages usually have a "References:" line which contains a list of globally unique message identifiers which allows a newsreader program to backtrack through the chain of discussion. Normally, discussion threads have a tree structure, as different people comment on the same article and thereby introduce a branch. But when you're posting an article, with most operating systems you can get at the draft of a message with a text editor before it leaves your machine. So: create two messages at the same time, but don't send them straight off. Instead, open both in a text editor. You then paste the message ID of each message (which will already have been inserted into the "Message-ID:" header line by your newsreader agent) into the "References:" line of the other. Then let your news posting engine send them both off.

 

A lot of newsreaders will try to build a tree of threads for a newsgroup when they open it up. The obvious way to do it is by assuming the "References:" line actually does encode a tree, or at least the last few nodes of it, searching recursively for linked messages. Oops. With mine, the recursion would never terminate. The result is that any newsreader with insufficiently paranoid checks for this situation would hang when opening any newgroup with one of my boobytrap message pairs. I know I clobbered Xnews that way, I'm not sure what others. Those messages are still sitting in Google's archive waiting for unwary software to step on them.

 

(You could get the same effect by having a message directly refer to itself, but I figured some newsreaders might be wise to that - using a pair of messages means there is no checkable local information to indicate a problem).

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...