At a previous job I was a developer on a medical instrument that used Windows to run the UI.
Before everyone gets all up in arms about it, Windows/Linux UI & database with external microcontrollers handling real-time control is a very common architectural choice for medical and industrial equipment. To the point where many Systems-on-Module (SoMs) come with a Linux-capable ARM processor and a separate, smaller processor for real-time, linked via shared memory.
Anyway, a customer called to report a weird bug that we couldn't resolve. After remoting into the instrument, we discovered that one of the lab technicians had attempted to install Excel on it. At some point the install must have failed, but it left a .dll behind that was causing a conflict with something in our code and keeping the UI from starting properly.
No, we did not learn anything from this incident...
Isn't this what Embedded Windows was always for, like for use in medical equipment, ATMs, POS, PLC, oscilloscopes, etc? Basically stuff that's supposed to be fire-and-forget, run 24/7 and that the user shouldn't be able to tinker with.
And also what group policies were for, that can disable the user from installing any software?
Am I wrong to assume that the fuckup was on your end, for using the wrong tool and not configuring it properly?
> Am I wrong to assume that the fuckup was on your end, for using the wrong tool and not configuring it properly?
Not at all. I agree that it should have been locked down and only privileged accounts should be allowed software update. But the system auto-booted into an Administrator account so it really wasn't a surprise that eventually someone would do something stupid.
I will say that this was for Windows NT retail, not Windows NT Embedded. At that point, getting an NT Embedded license practically required sacrificing your firstborn child. It was only when Microsoft got to Win XP Embedded that the license didn't look like it was written by a team of lawyers who already knew that they were perpetually in Hell.
Memories now of what we were given at the hospital long ago: our obstetrics ward was using Philips OBTraceVue software. The original FDA-approved system required Philips to package the OS and hardware all together, so we were given a bunch of generic Compaq desktops to run their fetal heartrate monitoring on.
The biggest annoying complaint was "we want to run our EHR software on it!" but because of the FDA requirements, we weren't allowed to install anything on the box. Yet somehow providing AV could be OK in some cases, and in other cases installing Citrix? And then we'd somehow find out someone managed to install the EHR client onto it anyways and it became a big old mess to have to have Philips come send a tech out of their own to reimage a PC we couldn't "legally" service.
It was a big messy pain for a while back in the day. Was happy when we finally got to upgrade to the newer IntelliSpace software on our own PCs in the ward. (Also got to meet a support engineer that came out rocking an Agilent badge, so that was super cool on its own right of history...)
> somehow providing AV could be OK in some cases, and in other cases installing Citrix?
The only way this could possibly have passed FDA scrutiny would be if the original manufacturer had validated this particular system configuration and approved it.
There's probably tons of stuff like this going on all over the place, but it manages to say under the radar, so no one notices it. But with the FDA's increased scrutiny on cybersecurity it will eventually disappear.
Back in like early aughts I remember seeing an ATM in Rome that had evidently crashed and was sitting at a DOS prompt. I was much younger then, but I remember thinking it wasn't terribly surprising, but it was also a bit of a wizard of oz moment.
since the astronauts are asking about it i'm guessing it wasn't snuck onboard. Sneaking stuff on to spacecraft to play with on the moon was a thing, i think one of the Apollo astronauts smuggled a golf club and balls to hit on the moon.
(I realize this mission is to only orbit and not land on the moon)
In a practical sense, I'm primarily interested in small to medium sized models being open. I think that might be common sentiment.
However, my hope is that there will be at least somewhat competitive big and open models as well, from an ethical/ideological perspective. These things were trained on data that was provided by people without their consent, so they should at least be be publicly accessible or even public domain.
You linked me to this comment from another one and I have to agree with this sentiment.
Creating these mini DSLs is something that requires a lot of thought and good design. There is a danger here as you pointed out sharply.
But I have some caveats and counter examples:
I would say the danger is greater when using macros and far less dangerous when using data DSLs. The Clojure community has been moving towards the latter since a while.
There are some _very good_ examples of (data-) DSLs provided by libraries, such as hiccup (and derived libraries), reitit, malli, honeysql, core match, spec and the datalog flavor of Clojure come to mind immediately (there are more that I forget).
In many cases they can even improve performance, because they can optimize what you put into them behind the scenes.
That's fair if you're looking at it from a performance perspective.
Not entirely fair if you look at it from a perspective of wanting fast feedback loops and correctness. In Clojure you get the former via the REPL workflow and the latter through various other means that in many cases go beyond what a typical type system provides.
> the opposite of boring
It's perhaps one of the most "boring in a good way" languages I ever used.
Taxes in Switzerland are generally quite a bit lower than in other European countries. However mandatory health insurance is not included as a tax in Switzerland and is one of the highest expenditures for regular folks and yes, cost of living is also generally very high. Specifically in urban areas (Geneva, Zürich etc.) but also in tax haven cantons (Zug, Schwyz etc.), because you pay a premium for housing/rent there.
Child tax credits are varied as well and generally too low (IMO), maternity/paternity leave and especially child support are much weaker than in Nordic countries.
From my perspective the strongest case for Switzerland is the decentralized and half-direct democratic process, political stability due to consensus. But you only get that as a citizen, which is not easy to achieve (takes a decade and you have to jump through a whole bunch of hoops). But the Swiss are very happy with their governmental structure in comparison and cherish direct democracy especially.
I had an idea about writing something similar, but for multimethods, but never got around thinking it through and trying it out.
The way defmulti and defmethod work is that they do a concurrency safe operation on a data structure, which is used to dispatch to the right method when you call the function.
My hunch is that it should be possible to do something similar by using core match. What I don't know is whether it's a good idea or a terrible one though. When you're already doing pattern matching, then you likely want to see everything in one place like with your library.
5 minutes ago | parent | next | edit | delete
reply