Congrats with the launch.

How does it compare with OpenShifts operator hub?

all operator hubs on openshift are backed by the actual projects operator, ie https://github.com/apache/flink-kubernetes-operator

> COLL 110 was a standard lectures-and-assignments college class - I lectured during our scheduled meeting time, then students did their projects on their own. Having tried this, I think that this is just not the future of education. This mode of teaching is designed mostly for the lecturer's convenience, but it's a terrible way to foster student understanding.

In this section I wish you would have mentioned the cost of teaching. Lectures might not be the best option for engaging students, but it’s quite effective when considering you can teach hundreds of students with one teacher. If 50% learns, that’s a good outcome.

I believe the fallout from lectures and university in general are just a part of the design. Not all people have the right motivation and opportunity to succeed at the university. One should work out how to maximize opportunities.

I've never been to a university that didn't run tutorials or practicals alongside lectures. Given a flipped classroom doesn't have lectures it can only be cheaper than the standard method, since the tutorials exist in either model.

> I believe the fallout from lectures and university in general are just a part of the design.

Just no. There are so many factors that go into students dropping out, many of which are outside their control (e.g. needing to support parents). There is a lot of research on this.

If $PROGRAMMING_LANGUAGE = go, you might be looking for https://github.com/containers/storage which can create layers, images, and so on. I think `Store` is the main entry: https://pkg.go.dev/github.com/containers/storage#Store

Buildah uses it: https://github.com/containers/buildah/blob/main/go.mod#L27C2...

Edit: buildkit seems to be the same, used by docker, but needs a daemon?

…and help the customer reconnect all devices on the WiFi?

Yes. It would be the same as resetting your email password and needing to login again on your devices.

If a password is so precious that you share it plaintext with third parties it is a bad usecase for a password.

The level of effort and obviousness of an email reset is nothing compared to helping someone figure out how to reconfigure every smart device ever made.

So it's a bad usecase for a password, then. Perhaps every router should ship with a preconfigured VLAN for shitty smart home stuff that is a lot more open, or maybe we should stop trying to stick internet into everything ever created.

Why should it be just the IoT devices that get the insecure network? Why not just stop trusting the LAN altogether and instead use technologies like HTTPS and DoH to ensure privacy on the important devices? That seems to be the way the tide is turning anyway.

Personally I'm all for that but people & packages seem to be pretty promiscuous about listen address defaults and assuming everything behind a routers NAT is trusted.

Treating the network as untrusted is good but as long as some people are paying for service, traffic and bandwidth there are reasons to not allow anything to use your network. And there is also a legal question of liability if someone is not quite above board from your IP.

I've actually worked help desk for about 3 years.

I've had calls lasting over an hour helping customers configure their email on their phone and computer.

I learned not to laugh when people called "the internet" either "that e-thingy", "mozarella foxfire" or "googlé charome".

I dealt with explaining to people why IE6 did not understand SNI when we decided to give all our customers websites HTTPS.

Just saying that I've been in that and seen that.

They can change it back after logging in if they insist.

they forgot the password, so they can't

Right, good point. There is of course the option to see saved wifi passwords on most devices... but I can see how an engineer decided to bypass all this bikeshedding and just send the damn password haha.

There's always the reset to factory defaults button. The vast majority of WiFi users have never adjust any of the settings anyways.

What happens if the next program in the pipe is not jb? Does jb also exit with a code?

For example `jb | jq`, where jq or a similar program discards the cancel character.

(Away from pc, unable to check right now.)

Good question! Yep, jb exits with non-zero:

    $ jb size:number=oops; echo $?
    json.encode_number(): not all inputs are numbers: 'oops'
    json(): Could not encode the value of argument 'size:number=oops' as a 'number' value. Read from inline value.
    ␘
    1

If you pipe the jb error into jq, jq fails to parse the JSON (because of the Cancel ctrl char) and also errors:

    $ jb size:number=oops | jq
    json.encode_number(): not all inputs are numbers: 'oops'
    json(): Could not encode the value of argument 'size:number=oops' as a 'number' value. Read from inline value.
    parse error: Invalid numeric literal at line 2, column 0

    $ declare -p PIPESTATUS
    declare -a PIPESTATUS=([0]="1" [1]="4")

So jq exits with status 4 here.

TietoEvry do the same in Norway, where accounts are prefixed with customer name.