Sorry for the follow-up delay. I think there’s no good way to do that except to trigger the certificate upgrade again directly. I know I had trouble doing it (”some certificates do that”), so the guy on the support side of things gave me a different URL to try. Turned out that it was the certificate upgrade URL, but I don’t have a reference to it right now.

I’m not sure that they still support access with the old kcrypto stuff. However, if you still want to try it, here’s the old one.

Hope this helps. I know even with access from Linux, the core ROS is still pretty dire. At least it’s really cross-platform now!

You don’t need it anymore. Once your account has been migrated to the new ROS system, the keys are converted and you can use the built-in Java plugin in 64-bit or 32-bit Linux. I’m assuming that this means it works as expected using OSX these days too.

If you still have trouble, I’ll try and help, but I’ll need more information about your system configuration. You can always reach me on twitter at @atownley and we can see if that will help.

Cheers,

ast

http://veryard.wikispaces.com/resistance

Fair point, but I wasn’t intentionally trying to restrict it. “Last” in this case can be interpreted as relative, so if you click “Back”, that’s now the “last” document as far as the hypermedia exchange is concerned.

The idea I was trying to get across was that you could fall over a hypermedia document lying on the street, pick it up and following the links should be all you need to do. I realize that this glosses over a lot about actually *understanding* the semantics of the markup, but I hope you get the idea.

One thing for sure that I DID NOT imply, was that you should embed “back” links of any kind in the document. You might have links that effectively go “back”, but only if it’s a valid state transition from the current representation. Only the app creating the representation can know this.

Thanks for reading the post!

ast

> The operations it performs should be totally bounded by the operations provided by the last hypermedia document received from the service, e.g. the service’s current application state.

Since there’s no guarantee, after whatever time the client needs to process the last document received, that it reflects the service’s *current* application state.

Therefore I’d relax this - the operations the client performs should be bounded by the operations provided by the hypermedia documents received earlier from the service.

I.e., the client can go back() in its history. Sure, the more back it goes, the higher the likelihood that the documents contain obsolete operations, but otherwise you would be suggesting that all hypermedia documents should contain explicit links back or up in the service resource hierarchy (back is hard because the service may not know where the client came from, modulo referer header), for example every item would have to link to (most) every container in which it is.

I think such design isn’t really necessary - the back button works quite well, after all, and there’s no reason it shouldn’t work in non-browser clients as well.

I like twitter as the mix as well. Some people I follow @simsron (Asst. Sec of HUD) for example, are data feeds. Kinda like an RSS feed from people I consider influential. Occasionally these turn into conversations, I’ll respond to a post with a comment or another link, which gets retweeted and the conversation continues…

Others, like my sister (@rboske) are completely a conversation and method for keeping a pulse on people.