[Gabriel54]

I'm surprised how few comments there are on this thread. This is probably affecting millions of students at the most stressful time of the year.

Incidentally I've always hated Canvas and probably every other LMS provider, but what is particularly amusing about this current outage is that it is occurring at exactly the time when universities are demanding that all professors put all of their materials on Canvas, without exception, due to ADA compliance regulations. It is explicitly forbidden for professors to, e.g., refer to pdfs posted on a personal website.

Other commentators here seem not to understand that many faculty also do not enjoy being forced to use Canvas.

[gchallen]

They have not succeeded in forcing me, yet. But it's sad how many computing faculty apparently can't operate the basic online infrastructure needed to support their courses. Not that universities make it easy for us.

And of course the other serious concern I have with Canvas is that they are likely using all the materials faculty upload to train their AI replacements. Many of my colleagues engage in dark humor about this but I haven't noticed much action.

[JumpCrisscross]

> they are likely using all the materials faculty upload to train their AI replacements

Instructure (Canvas's developer) partnered with OpenAI last year [1], about a year after KKR and Dragoneer (PE firms) acquired it [2].

[1] https://www.forbes.com/sites/rayravaglia/2025/07/23/instruct...

[2] https://www.pehub.com/kkr-and-dragoneer-complete-4-8bn-take-...

[lucas_v]

instructure/canvas-lms is open-source -- is there anything preventing universities from hosting it themselves?

[Meneth]

A bunch of plugins running on canvas.instructure.com are proprietary, according to their FAQ: https://github.com/instructure/canvas-lms/wiki/FAQ.

I would guess these plugins are chosen so a majority of user won't want to live without them.

It also seems these plugins "link" to canvas-lms, so keeping the proprietary would be a GPL violation if anyone except Instructure holds part of the copyright to Canvas.

[dotancohen]

Money, skill, liability.

That calculus is about to shift.

[FloorEgg]

I'm sure the engineers at instructure are not capable of building systems that can do that. You give them too much credit.

[freedomben]

Former Instructure engineer here. Ive been gone almost 10 years at this point, but some of the best engineers I've ever worked with were at INST.

I'm not sure where your stereotype even comes from, because Canvas is not trivial software. You can see for yourself as it's AGPL and I assume you looked at the code before criticizing it because any good engineer would do that.

[w4der]

I don't care how good you think it is, the fact that it (back when I used to be a TA) would break if two TAs tried concurrently grading different parts of an assignment of a student is bonkers. The workaround for that was to use a Google Sheet document so TAs just looked at the submission in Canvas, then filed in their grades and feedback on the sheet. The issue is that Canvas, as far as I could tell, did not support mass uploads from a csv, so we had a script which would read every entry on the csv, map that to the student's ID and grade them, which made it look like the TA which had generated the API key graded all of the students (and would get all the backlash from poor grades).

I completely agree that it is not trivial software in the worst sense, it tries to do too much, while not being particularly good at any one of those things, and is way too rigid for how diverse the needs of different courses might be even inside a single faculty. And saying "It's AGPL, just self host and add your requirements to it" is not really useful, that would mean way more money and effort than what a university's overworked IT dept. is capable of.

[slfnflctd]

This particular section of your comment sounds a helluvalot like some of the SAP implementations I've seen:

> it is not trivial software in the worst sense, it tries to do too much, while not being particularly good at any one of those things, and is way too rigid

[FloorEgg]

I didn't say instructure engineers were bad.

What I meant is they aren't capable of building AI capable of replacing professors. I still consider it a reasonable assumption, as it has nothing to do with how well engineered canvas is. It's a different competency than instructure would have, and I've heard from insiders instructure has been spinning their wheels on way more trivial AI challenges. I also understand well how hard it would be to create AI that replaces professors and how the current best AI from Google, Anthropic, OpenAI is orders of magnitude away from being able to do that.

An engineering culture can change a lot in 10 years, and a company's engineers' ability to do stuff depends both on the individual engineers abilities as well as the company systems and culture.

[mbesto]

> some of the best engineers I've ever worked with were at INST.

> You can see for yourself as it's AGPL and I assume you looked at the code

Can you look at any codebase and tell me it's written by some of the best engineers and it's not trivial?

[hackyhacky]

I've been using Canvas for years and it's some of the worst written software I've ever used. It's slow, buggy, with an atrocious 2001-era UI. It's a CRUD app that has no excuse for being so cumbersome. I'm not surprised at all that their security is just as bad as the rest of the product.

A bright undergrad could build a superior replacement in a few months, even without AI.

[freedomben]

I won't disagree on usability. It has some sharp edges for sure. But

> A bright undergrad could build a superior replacement in a few months, even without AI.

Is quite naive. Canvas is not at all just a crud app. You can view the code yourself as it's AGPL

[ericpauley]

What component in particular goes substantially beyond CRUD?

[hunter2_]

If they're at the level you say, they just might install some AI gizmo like the Vercel employee was accused of, but really let it run amok with write permissions.

[bradley13]

I'm in Europe, and we don't use Canvas (at least, I've never heard of it). However, we have similar diseases. In my particular school, it's a massive SharePoint site plus ever more stuff in Teams. Plus Moodle, plus other services.

The MS services have not improved teaching at all. What they do, is fragment communications, and add ever more places people have to look, in hopes of finding things.

But the administration loves them. "The bureaucracy is expanding, to meet the expanding needs of the bureaucracy."

[pfortuny]

Spain here. Most of our public Universities have their IT stack on MS... I cannot fathom how much of our national budget goes to their pockets.

Thankfully, I store my teaching materials on my personal non-uni webpage, and the student's marks in my office's computer (apart from the MS-based Uni system).

Whenever something happens with MS, chaos ensues throughout the whose Uni and the students end up paying the consequences.

[iamflimflam1]

Teams and SharePoint eventually infect any organisation that uses Office.

[tokai]

There are plenty of European Canvas customers.

[onetimeusename]

Live streaming of class through Canvas is very popular. Quite a few people just watch from their dorms. So maybe people will have to come back to class, that will be entertaining. The class rooms are almost standing room only (sometimes they are) on the first day of class and then gradually thin out. Sometimes 10 or so people show up out of a class of 100. If Canvas is not back up soon I think it could actually be disruptive for that reason also.

[ecshafer]

This is awful to hear. The idea that students are just half assedly streaming the lectures is really just ruining things in the long run. This is a bit old manny, but showing up to lectures is good. You go to class, you get face time with professors, you can ask impromptu questions, you rub elbows with classmates, you talk on the walk between classes, you maybe run into a cute girl. Friction like walking to class and finding a nook in that annoying hour gap you have, are the things that make life enjoyable.

[gwerbin]

When I was in school, professors attitudes around attendance was usually "you're only hurting yourself, I don't care if you show up or not".

It's been long enough that I can't claim to be in touch with the current generation of teaching faculty. But it might be an element of that, combined with the desire to provide accessibility for the handful of students who do in fact need the accommodation.

[tokai]

Showing up to lectures is vastly overrated. Like note taking it's cargo cult behavior for middling students that care more about going through the acts of studying, than actual learning.

[ecshafer]

You'll note I didn't mention quality of education in my arguments. I am talking about the human experience. Though the studies typically show a correlation with grades and attendance.

[timdiggerm]

What a failure of university leadership to allow or even encourage that practice

[altairprime]

Not much overlap between students and HN these days, though? I’m an extremely rare outlier afaik :)

The administration has so far opened with one “Canvas said” and then an hour later one “Canvas is down indefinitely” email noting that they’re aware it’s serious.

(Canvas is a glorified wiki for teaching students, with quizzes and such, for those unaware.)

[dang]

> Not much overlap between students and HN these days, though?

That's my biggest fear.

[byronsharman]

I'm an undergrad student in computer science and I come here regularly. Many of my friends do the same. Of course, that can't be extrapolated to students globally, but students who love what they do are not extinct!

[dang]

You guys are making me feel better.

Same question for you as https://news.ycombinator.com/item?id=48065589, btw: what do your friends read besides HN?

[byronsharman]

HN is kind of the go-to platform. We also have a Discord server where we hang out and talk about stuff that may or may not be software-related. I know some of us are interested in getting accounts on lobste.rs, but we need to find someone to refer us. I keep forgetting to ask one of my professors.

[gucci-on-fleek]

FWIW, I'm a student, so there are at least a few still here. Feel free to ask me any questions (either via email or via replies to this post) and I'll try to answer them.

[dang]

Ok! How many of your smart friends read HN? and of the ones who don't, what do they read instead?

[gucci-on-fleek]

> How many of your smart friends read HN?

I don't think that any of them do, but I'm a Canadian math/physics major, which is slightly outside the target audience for HN.

> of the ones who don't, what do they read instead?

For the social aspect: mostly medium-sized Discord servers. For the news aspect: nothing at all. Both of these do have some advantages, but it's still a bit of a shame, because the Discord servers aren't indexed by Google, so they're hard for outsiders to find, and not reading the news means that they're missing out on some of the cool new tech advances.

[strix_varius]

Is there any internal data on where students are going instead?

[dang]

Not much, but I do ask the youngest founders what their friends read if they don't read HN, and the only consistent answer I hear is Twitter.

(and btw, they do say "twitter")

[AuthAuth]

Many of my sisters friends do everything entirely via tiktok. They look at what trends are popular and they target that fully on platform. This is for stuff like building niche targeted apps, selling beauty products/clothing brands, restaurants.

[DaSHacka]

You honestly don't wanna know

If my peers are any indication, a whole lot of TikTok, Reels, Twitter, Discord, and other such mind-numbing platforms.

The types of platforms I would consider 'substantive' (or, at least, more substantive than those platforms) are definitely on the way out.

The few times friends have seen me browsing Hacker News or a certain Mongolian basket weaving form, the first thing they comment on is how confusing the interface is, and how old the site looks.

I truly don't understand the mentality, but if your site doesn't take three seconds to buffer a simple text drop down menu, and have JavaScript elements load in mid-scroll that bump elements around the page making you just barely miss that button you were trying to click, then your site is seen as 'inferior' or 'sketchy'.

Perhaps I've just had a bad sample, but I've experienced a variety of different environments by this point, and by and large, I've seen more people in my generation act in that manner than not.

[dang]

This is actually reassuring. We don't need all your peers! We just need you and whatever smart cohort you're bonded with.

It's true that HN looks old - it looked old before you were born, probably - but (a) I have no idea how to change it, and (b) HN is a long bet on plain text. If the smartest young people lose interest in reading, I'm ok with HN dying for that reason. I just don't want it to die for any cheaper reason.

[gsquaredxc]

I would like to offer some additional reassurance: I send my friends articles I see on HN that might interest them. A (in my view) very good litmus test is when someone asks where I saw it, because this demonstrates some desire for continual learning. I find that anyone that asks that question seemingly trusts an interface like HN more because of it. My suspicion is that this is probably because at a certain point you see stuff like Agner Fog's work, LWN, or a number of other minimalist websites and realize that a website that is popular despite the lack of overindulgence in UI must be popular because of the content. It doesn't hurt that the best courses in my university experience have had websites that have not changed much since the late 1990s (one did change the lime green text on turquoise background on their page after the recession to a color scheme that didn't cause headaches in students).

I do find that my peers that now read HN used to be judicial about curating a Reddit feed and mostly otherwise limited on other sources. Short-form content is addictive and as nearly as unavoidable as sugar, but many of my brighter peers work on reducing that intake. Long-form YouTube is also something I find to be a marker of someone who is seeking knowledge. Many of my peers do scroll Twitter and TikTok all day, but I find that those who are easiest to chat with are those who have already scrolled HN today and want to discuss a particular article they know I would have seen. I've had conversations that start with "Did you see that TikTok?" and conversations that start with "Did you see that article on HN?" and the latter is always more engaging.

[tailscaler2026]

Discord is just chat, I wouldn't call it mind-numbing, reminds me perfectly of IRC from a utility perspective.

That said, it's a commercial closed-source single point of failure.

[Kiro]

How is Discord mind-numbing?

[DaSHacka]

The massive discord servers have a culture that is no different than any other social media

[Ronsenshi]

Perhaps some interest-related Discord servers. Tragically, Discord is just another locked down silo without publicly accessible front on the web.

[daedrdev]

I think its a good fear to have, I feel like many sites dies when the main path of discovering them broke for one reason or another, who knows what the path to discovery of this site is would be for a student today.

[altairprime]

Drop me an email if you like — it’s not really topical to Canvas but I’m happy to discuss further.

[dang]

(Comments were split across multiple threads and we've since merged them.)

[Gabriel54]

Definitely not a criticism of your (hard) work here. Thank you!

[dang]

Thanks! I just added that bit to pre-empt confusion - context-switches like this are one of those rug-pulling moments (https://news.ycombinator.com/item?id=48041875).

[MarsIronPI]

We all appreciate the work you do! Thank you!

[apublicfrog]

Can you explain for the billions of the rest of us why this is the "most stressful time of the year" for the group you're referencing? I assume that's American students and/or teachers?

[isakmarr]

Final exam season, and it's ongoing in Iceland too, so not just American students.

[Steve16384]

Here in the UK it's currently exam season. One of my son's had a GCSE exam just today.

[pfortuny]

European students are preparing for their finals.

[cocoto]

Replace your material content with lorem ipsum or garbage LLM content and upload it to Canvas to test the accessibility of your documents if required.

[isityettime]

What? What makes Canvas accessible in a way that HTML and PDF files are not? It's true that PDF readers aren't the best for screenreaders, but surely you can just upload a .html copy as well.

[Gabriel54]

Canvas has an easy way of checking if a pdf or other course material is accessible, so many universities are forcing faculty to put all their materials on Canvas. That way if a pdf or powerpoint is not compliant it is immediately flagged. The goal is to reach a "100% accessible" metric.

Note that little of this really helps the students that it is supposed to help, because as you wisely point out, raw HTML is almost by definition extremely accessible. I work in a field that uses Latex and the source code of Latex should also be considered more accessible than the compiled pdf. But for university administrators the only important thing is that the accessibility metric that appears (or used to appear, before today!) on Canvas shows 100% accessible.

[isityettime]

That really sucks. I'm visually impaired and many members of my family are/were blind. I think accessibility is really important, but it's so painful to me to feel like people's limited energy is being directed towards performative measures, useless rituals, vanity metrics, etc.

Nobody has infinite energy, and disabled people don't have infinite social capital. It's a shame when energy from that shared pool gets spent on things that don't really impact meeting people's access needs.

And the other thing is that everyone's access needs are different. It can certainly be useful to try to set a baseline or propagate common guidance. But the most important thing, especially in a university setting, is for instructors to be flexible and responsive and for classes (and non-teaching workloads) to be structured in a way (e.g., small enough) that supports that.

I think metrics like "100% accessible" might even be dangerous. It makes it easy for able-bodied people who aren't in direct contact with disabled stakeholders to pat themselves on the back without actually knowing what's going on.

Bleh. Good luck doing right by your disabled students and disabled colleagues, and good luck resisting the bullshit.

[Gabriel54]

I was only a lowly TA so I saw these issues from afar, but I would add that, on a more optimistic note, I don't think I've ever met an instructor who wouldn't do whatever he or she had to do to support someone with special needs. As you suggested, metrics do not tell the whole story and certainly metrics for the sake of metrics are not helpful and may in fact be dangerous.

That said there is certainly a lot more work that needs to be done in this area. Hopefully these regulations over time bring out practical positive change. Time will tell.

[bradley13]

Why does everything have to be 100% accessible?

I'm a prof. When I have a student with special needs in my class, the administration tells me ahead of time. I make the necessary allowances - and those differ from case to case, anyway: whether it's extra time in exams, or someone who is deaf, or someone who is blind, or whatever.

When it happens, I make the necessary allowances. When I don't, then...I don't.

The obsession that everything has to be 100% accessible, for every kind of disability, all of the time? That's just nuts, not to mention a complete waste of resources.

[altairprime]

The attitude they’re contesting is that accessibility is a “minimum compliance” category: people tend to invest zero effort into accessibility until caught, and enforcement that waits for students to report suffering is terrible, so automated analysis of accessibility that is ‘always on’ dramatically raises the water level for all accessibility. It won’t reach 100% accessible but it’ll reach a lot higher than the 1% accessible it was otherwise, and that’s a valuable result worth obsessing over. Doesn’t have to be complex: “Your video was uploaded without captions”, “your PDF is missing a text layer” are probably the two most valuable and simplest to implement rejections on the table.

[Telemakhos]

Universalizing statements like "100% accessible" are usually bad ideas. In this case, it's driven not by administrators but the Department of Justice, which is rulemaking accessibility via consent decrees. I think a lot of people miss that and just blame the administrators. Rulemaking is a long process, and the rules being made are stuck in a time before AI could reliably read a book to a blind person: the rules shift the onus onto the content creators, when we've created a whole new ecosystem of ways to eliminate the onus. The DOJ should probably step back and stop trying to regulate this, because the market has already solved it.