Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Somehow this reminds me of a problem my father (who is a doctor) has often complained about. Basically IT shitheads enforcing their policies on ALL computers in the hospital, including those used in operating rooms.

So that means things like a screen (displaying important information) locking from X minutes of inactivity, on a computer that the surgical team is physically unable to "bump" periodically or type a password on, due to them being scrubbed and sterile.

It reaches a point where you basically have to tell these people "Someone could DIE if you don't change the fucking policy for our use case" to get things to change.



Conversely, computers outside the surgery are all on one big network in a public space with random unsupervised people walking around at all hours. It's very easy for someone to utilise this in a way that violates the privacy of a patient in the worst possible way. Think nosy reporter looking for HIV status of a celebrity, or abusive parents looking for their kids to stop them having an abortion, that kind of thing.

I've been involved in the design of hospital computer networks, and I tell you: meeting all the requirements at once is hard!

The system we designed used contactless smart cards and Citrix. The idea was that as the attending doctors moved from bed to bed and ward to ward, their desktop session would move with them.

The instant they logged on somewhere else, the previous terminal would lock and the session would transfer to the new terminal without a full Windows logon cycle. It was basically equivalent to disconnecting a monitor and connecting a different one. No passwords were needed, they just had to tap their id card once.

My challenge was that this has to occur in under a second, including the smart card cryptographic authentication step, which was limited by the throughput of the NFC chip on the card. From memory, it was woefully slow, and we had to use the smallest compatible elliptic curve cipher available to make it acceptable.

Similarly, it was difficult finding a thin terminal device that was both fast enough to do this, and fanless so that it could be sealed against dust. This was needed to prevent their warm insides becoming the perfect breeding ground for antibiotic resistant superbugs.


This sounds like EXACTLY what Sun's SunRay thin clients used to do, back when the rest of the *nix world seemed completely oblivious to this whole "hot-desking" concept.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: