Hospital to Run GNU Health, OpenSUSE | Hacker News Reader

Hospital to Run GNU Health, OpenSUSE

news.opensuse.org

41 pointsCTOSian9 comments

The handful of Rasberry Pi SBCs running this are probably capable in terms of computing resources, but the lack of RTC would give me pause. If a power outage occurs and the hospital comes online before the ISP (many hospitals use generators for such events), then no NTP will be available to the Pis and their clocks will be messed up. That could impact HTTPS, which is important in this context. Hopefully there is a UPS in place that will keep them online until the internet comes back up.

It is not hard to run an NTP sever locally. At a hospital, I would expect that such a server would be supplied by both a UPS and a generator, and it would have a battery-backed RTC.

I would be aghast if the RPi were directly connecting to anything outside of the local network. That would be a major failure of security in my book.

Cool. I love openSUSE.

>Patients can access personal health records through MyGNUHealth running openSUSE on the PinePhone; or running KDE’s Plasma Mobile, XFCE or their favorite desktop environment

Oh no. How available is the PinePhone in Cameroon? Also, openSUSE builds for the PinePhone have been broken for months

That just seems like wishful thinking. The Pinephone is barely usable as a phone - it's market penetration in Cameroon must be essentially 0. And what percentage of people have a "favourite desktop environment"?

I suspect if people have smartphones in Cameroon, they're overwhelmingly Android ones. Unfortunately it seems that MyGNUHealth doesn't support Android.

Hopefully US hospitals follow the lead and get rid of that dumpster file known as epic

Epic isn't perfect but most customers seem to like it. Are the alternatives really any better? Most of the problems seem to be caused by customizations or lack of training rather than inherent flaws in the product itself.

https://www.beckershospitalreview.com/ehrs/klas-ehr-market-s...

The problem is Epic's mentality of "do it our way Or Else" which seems to be across the entire software package.

I once asked my former hospital IT team why we didn't get Epic ourselves and it was described to me as "You don't get to buy Epic, they reach out and 'offer' to sell it to you". You then buy what they tell you to buy, hire who they tell you to hire, train what and who they tell you to train. If you do any bit of it wrong it eats into your support costs because of a "quality of install" score that either raises or lowers your annual support fees.

That's a reasonable position for Epic to take. Most failed EHR implementations have been caused by excessive customization. Hospital staff often wanted to make the new software fit their existing idiosyncratic processes and forms. But in reality there was no valid business or clinic reason for most of those differences between hospitals. It's better in the long run to reengineer hospital operations to match industry best practices and the selected EHR, regardless of whether that EHR is Epic or one of their competitors.

As a local hospital IT director told me, "Vanilla is the best flavor." By that she meant that if they purchased a piece of software they were going to use it as-is and not get sucked into a black hole of customization.

Atleast epic's MyChart runs on Android and iOS.

Reply on news.ycombinator.com