| ▲ | e28eta 7 hours ago |
| Students having records of what their score was doesn't prove to the professor / university what score they received. "FWD: Exam 1 Results" is not especially auditable. |
|
| ▲ | lacunary 7 hours ago | parent | next [-] |
| If only we had some way of signing messages |
|
| ▲ | JumpCrisscross 6 hours ago | parent | prev | next [-] |
| > Students having records of what their score was doesn't prove to the professor / university what score they received It's better than nothing. (And good training for the real world.) Also, most universities (and many schools now) issue academic e-mail addresses to students. In those cases, the email is definitive proof. |
|
| ▲ | AmblingAvocado 6 hours ago | parent | prev | next [-] |
| DKIM signature could be used to verify that Canvas' server sent the email with the given content |
| |
|
| ▲ | gruez 7 hours ago | parent | prev | next [-] |
| As opposed to a screenshot of a website? Presumably the professor has a spreadsheet of all assignment grades that is submitted to the school? |
| |
| ▲ | JumpCrisscross 6 hours ago | parent | next [-] | | > Presumably the professor has a spreadsheet of all assignment grades that is submitted to the school? This would undermine Canvas's lock-in. | | |
| ▲ | freeopinion 5 hours ago | parent | next [-] | | Canvas is built to automatically export its gradebook to an external system. It will do that automatically every day if you want it to. Teachers or others can manually export to the configured foreign system on demand. So if you grade something and want it to show up in the foreign gradebook without waiting for the daily export, you can just press the button to make it happen right away. | |
| ▲ | doctorpangloss 6 hours ago | parent | prev [-] | | i cannot believe how much benefit of the doubt people are giving canvas ed tech is the WORST performing VC sector the ONLY game in that town is vendor lock-in! are people joking? c'mon, canvas is a huge piece of shit. the SaaSpocalypse is coming for them - it seems it is simply that LLMs will be used to exploit it first, rather than universities writing an open alternative they share with each other for free. | | |
| ▲ | freeopinion 5 hours ago | parent | next [-] | | Canvas is AGPL licensed. Moodle is GPL. Universities or anyone else can already contribute to big name LMS. Canvas is used by Harvard, MIT, Stanford, Carnegie Mellon, CalTech, etc. If they each paid 10 FTE, they could set up a foundation that could govern the development of a top-tier LMS. Every tier-1 state institution could contribute 5 FTE. Even little JuCos could chip in an employee here and there. You'd pick up hundreds of capable employees at a fraction of what those schools currently pay to Instructure. | | |
| ▲ | freeopinion 5 hours ago | parent | next [-] | | How well has this worked for Open edX? | |
| ▲ | gizajob 2 hours ago | parent | prev [-] | | Why do they all pay for it then? Seems pretty universal in the UK too. Is it having the benefit of someone to blame when things go wrong? |
| |
| ▲ | freeopinion 5 hours ago | parent | prev | next [-] | | On paper your idea seems obvious. You take a bunch of institutions that actually teach students how to program and have them cooperate to build an open LMS that benefits them all. In reality, universities always spin off anything that looks like it could generate revenue. It is very telling that you can't even get your college transcript from your college. You have to go to (and pay) some third party to get it. Some universities even outsource their "classes" like elderhostel to cruise lines and travel companies. | |
| ▲ | gucci-on-fleek 5 hours ago | parent | prev [-] | | > rather than universities writing an open alternative they share with each other for free That already exists [0], and is actually reasonably popular. > the SaaSpocalypse is coming for them - it seems it is simply that LLMs will be used to exploit it first I doubt it, because enterprise sales has nothing to do with how good your product is, how expensive it is, how easy it is to administer, how secure it is, etc.; it only depends on how good you are at enterprise sales. I mean, my university is Oracle-based, and I'm pretty sure that you could get 3 random undergraduates to write something better, so I don't think that LLMs writing better/cheaper software will make any difference here. [0]: https://moodle.org/ |
|
| |
| ▲ | blahedo 7 hours ago | parent | prev [-] | | Nope! We're encouraged to keep all that exclusively in canvas. (As noted, I have my own spreadsheet. But I'm an outlier.) |
|
|
| ▲ | gucci-on-fleek 5 hours ago | parent | prev | next [-] |
| Presumably the system will be back up eventually, so there's not much benefit to lying here, since at best you'll raise your grade in a few classes for a couple months, while taking on a pretty big risk of getting caught. |
|
| ▲ | pishpash 7 hours ago | parent | prev [-] |
| You forget things can be signed, with the key owned by the school. It can be done. |
| |
| ▲ | SlightlyLeftPad 6 hours ago | parent [-] | | Does signing really make this easily auditable from the professor’s perspective? | | |
| ▲ | DaSHacka 6 hours ago | parent [-] | | Exactly this, when was the last time a HN user had to interact with the prototypical 60-year-old set-in-their-ways professor? Extremely non-tech savvy, hates computers, and is gonna grumble "What the hell is a PGP? Better not be another one of those phone code things." as you try to pitch this highly-technological solution to a largely niche problem domain. | | |
| ▲ | jazzyjackson 4 hours ago | parent | next [-] | | I mean a cloud based learning management system also seems to be a very technological solution to the very old problem of checks notes grading quizzes? | |
| ▲ | Forgeties79 6 hours ago | parent | prev [-] | | They don’t even need to not be tech savvy. This stuff just registers as “hassle” to most people so they do the bare minimum or search for ways to not deal with it at all. It’s easy to “tut tut” at them but ultimately we need to accept reality: privacy, security, these things take extra effort that isn’t strictly necessary for people to go about their daily lives even though the stakes can be super high. It’s not a problem until it is, so they aren’t really barriers that require people to do the work. It’s like convincing someone who just simply doesn’t want to go out and buy/install a lock on their door to go do it, except it’s not even a one-time thing. Their door works fine. They can come and go as they please. It’s not until something happens that they maybe change their tune (and even then!) Hell just getting people to do secure passwords is a whole thing. |
|
|
|
