@48 I doubt scratch is taught at university level. That is what quorum looks like to me, an easy language with accessible tooling for kids. If they try to make it become more than that, they are mistaken.
You are not logged in. Please login or register.
2024 Note on Registrations
the old system to register for an account is back with a few security improvements. Please tell your friends to click the register link and, once a password is confirmed via email, visit the sticky topic in the "Introduce Yourself to Gain Access" room.
Is quorum a good place to learn how to coad (Page 3 of 3)
AudioGames.net Forum → Developers room → Is quorum a good place to learn how to coad
Posts: 51 to 58 of 58
#52 2021-05-06 04:16:52
@50
That's what happens when you unknowingly ask about something that we all hate, beyond just the level of "I wouldn't use it because it sucks" and into "the world would be better if this didn't exist".
@48
What did the Linux people do? Accessibility or something else?
@49
I don't know about mandatory. Synthizer gets into some pretty serious stuff. Not maximally serious and maybe you've gone further, but serious enough. I'm not saying it's not useful if done right but no one has done it right, and that may actually be a technically insoluble problem that's roughly equivalent to AGI.
@51
Quorum kind of explicitly sells itself as something that wants to be taught at universities though. That's why we're complaining. Their mission statement isn't "and then you switch to something else". If the goal was that we give it to gradeschoolers and then we help those people get off it and onto something else--well, I'd still complain because of all the money, but at least it makes some sense and wouldn't be actively harmful. If the expectation was set that you do definitely have to leave this for serious programming yes really we mean it, then it wouldn't be taking away from anything else and the entire discussion here would just be "I guess it's useful but man we could spend that money better". But unless you've got sources or something that I haven't seen, that's definitely not what they're doing and their goal is to try to replace everything.
As for VS, 2010 is when it broke with Jaws for something like 3 or so years, and I did reach out to people, and the entire thing with respect to workarounds was "I happened to already know all these keys so it sucks but I'm making do, but you have no way of discovering them". So I have no idea what you were using, or maybe you just had the patience to randomly hit things and put yourself through the bug thresher from hell, but this broke in the middle of college and it wasn't fun. Maybe if it broke again now I could work it out, but I don't think anyone here will dispute that me not being able to use it despite really, really needing it counts as inaccessible.
My Blog
Twitter: @ajhicks1992
Twitter: @ajhicks1992
#53 2021-05-06 08:11:00 (edited by Ethin 2021-05-06 08:16:48)
@52, no, the Linux thing was completely different from what Quorum is doing. The Linux issue happened very recently. If you Google it you'll find quite a bit of info about it, but the executive summary is that some researchers at UMN though it would be a good idea to try to send in invalid/vulnerable patches into the Linux kernel to try to subvert the review process to see how it could be improved. They didn't ask for permission from the community because they knew they wouldn't get it, so knowing that they wouldn't get it they went ahead and did it anyway. The patches were supposedly rejected (though one patch designed to introduce bugs was buggy itself, making it a valid patch). However, I've read conflicting reports from the LKML that indicate that the patches did in fact make it into some stable Linux kernels. The results of all of that was that UMN got banned from contributing to the Linux kernel in future until they can demonstrate that their intentions aren't in bad faith; the kernel maintainers are now reviewing every patch submitted by UMN to verify that they weren't introduced in bad faith, and if they were, they're being reverted; and the reputation of the researchers and the university have been seriously damaged. Not to mention the Linux kernel community is seriously pissed off. The researchers were going to submit a research paper about it, but the Linux Foundation forced them to withdraw it. The reason I mentioned it in 48 was that that ties into what Quorum was doing: they're a bunch of researchers, and like the ones in the Linux kernel fiasco, they (1) had absolutely no idea what they were doing in terms of kernel code or code quality, but they knew exactly what they were doing in terms of harming the community and (2) they're deliberately and knowingly releasing a harmful product to satisfy intellectual curiosity. Now, I don't know if the Quorum devs are doing it deliberately; they may not know that what they're doing is harmful. However, considering how long the project has been around, I'd find that very hard to believe. Someone has to have approached them and told them that they're harming the programming community.
Edit: for point (1) I mentioned kernel code but substitute appropriately. Both actions are harmful and I wouldn't buy it if Quorum came on here and said "We had no idea we were harming the community of people who are wanting to learn programming". If the project had only been around for a year I'd get it, but I think Quorum has been around for what, 5-6 years now, if not longer?
"On two occasions I have been asked [by members of Parliament!]: 'Pray, Mr. Babbage, if you put into the machine wrong figures, will the right answers come out ?' I am not able rightly to apprehend the kind of confusion of ideas that could provoke such a question." — Charles Babbage.
My Github
My Github
#54 2021-05-06 19:56:11 (edited by vortex1024 2021-05-06 20:04:33)
@52, ok. I read an article about the quorum people going at a school for the blind and teaching kids, and supposed that's all there is to it. Guess they thought their language can do better and tried to extend its scope. Their chanse of succeeding is less than \epsiolon, but everyone has the right to try. The ones who gave them the money mayy be flagged guilty, but we don't know what alternative projects they had on the table, so it is really hard to accuse someone, in my view.
The only viable project that didn't receive financing I know was your audiograph.
as for vs, I think I was using 2008 some time after 2010 was released since I liked the jaws scripts for 2008. When I decided to move, it worked. I think I jumped to vs2012
#55 2021-05-06 20:39:07
@53
I've spoken to them some years ago. They aren't doing it on purpose but they aren't particularly interested in criticism. This is mostly the typical science vs. engineering divide (e.g. ask a physicist to build a bridge and so on) just in a particularly harmful place for that divide to decide to end up being. That said, that was almost 10 years ago now and I suspect that they're continuing mostly because of momentum at this point.
@54
I'm well beyond caring about audiograph, and I only use the Holman prize as an example because it is something I'm personally familiar with. I would have liked to have seriously done that project, and maybe I will one day, but there's clearly very little interest. But Holman is a drop in the bucket.
To give you another one, though I don't have a source other than "I was at the guide dog school and asked", guide dogs for the blind gets 30 million a year and only gives out a few hundred dogs. There's inefficiencies and bullshit like that all over. I could go find you a bunch more if I really cared to. People don't give money to the most effective thing, they give money to the thing that makes them feel good inside and hits the warm fuzzy button. Quorum does that because it's helping blind kids learn to code. It doesn't matter what someone like me says, people will almost always choose warm and fuzzy over effective. People say their primary motivation to help is that they want to help, but really for a vast majority of people the primary motivation is to feel good about yourself, which is how we get into a disfunctional world where blindness-related money goes to the least effective places because those places happen to have a cool story or whatever.
I feel like your model of how funding works is too nice. It's not about effective at all. There are two factors: are you asking, and do you have the best memes? there's plenty of places to put the money. If you just line up number of projects asking for dollars and number of dollars available there's more being asked for than is available. But Quorum has a good meme so Quorum wins. It's as simple as that.
My Blog
Twitter: @ajhicks1992
Twitter: @ajhicks1992
#56 2021-05-06 20:54:58
Hey @camlorn, what do you think about Mathjax's built-in accessibility mode? In modern versions of MathJax, it seems to be much better than what it used to be. In particular, it does not try to interpret math in any way, but converts it to speech exactly as written. For example, instead of "limit as x goes to negative infinity of f of x", you get something like "l i m subscript x right arrow minus infinity endSubscript leftParenthesis x rightParenthesis". After being rendered, the equations are visible as plain text, so you can use your screen reader's word-by-word navigation to read them. I find it to be much more enjoyable than reading raw Latex, as it gets rid of all the fancy markup, extraneous braces and all that, but easily communicates information about letters being uppercase, bold etc, something MathPlayer itself rarely does. There's also a way to navigate by the tree structure, by focusing on the math and turning browse mode off, although admittedly I haven't used that feature very often. Mathjax's own accessibility documentation explains it pretty well, though.
The whole thing seems to be based off MathSpeak, which seems like exactly what you're looking for. It's literally "Nemeth but for speech", developed by prof. Nemeth himself. There are even brief modes, although it seems like MathJax doesn't expose them properly.
There are still bugs and edge cases you run into once in a while, particularly when MathML is (ab)used and structures intended for i.e. representing fractions are used for something else, but it's much less annoying than MathPlayer.
It definitely has problems, and the option to view the plain source of an equation still comes in handy once in a while, but for me at least, it's an improvement over anything else out there.
Speaking of other recommendations for people in general, Another productivity enchancer worth knowing about is NVDA's powerful support for speech dictionaries, particularly when paired with regular expressions. It works particularly well when interacting with your own Latex, or when you interact with one source of markup frequently. Fiddling with and abbreviating the names of punctuation symbols also helps. For example, I have "par" and "rap" for the left and right parentheses, "lace" and "race" for braces, "back" for backslash, "is" for equals, "score" for underscore etc. I also abbreviate common LaTeX and programming constructs, so an empty pair of parentheses becomes empar, for example. I abbreviate longer things too, Go's "if err := <something>; err != nil" becomes "earn <something>". Learning to navigate by indentation also helps tremendously.
NVDA used to be much better at this pre speech refactor, sadly. Breaking everything up into tiny little chunks really doesn't help for this use case.
#57 2021-05-09 01:05:19
@55, maybe my model is unrealistic, but still. Quorum received money since it was an universitary research project, right? I guess state funding is separated into categories, i.e. the money for universities is separate from the money allocated for private companies to equal the playing field for the disabled, so the only alternative for the quorum money was to go into another research project, right?
#58 2021-07-10 13:40:58
Just some anecdotal stuff: Quorum 3 was released a few days ago, so I thought I'd give it a try. I do kinda like their easily-understandable errors, but like, some aren't any better than Python's "syntax error." Also, the thing crashed on me, twice, once when running a basic file with only calls to use a some libraries... packages in Python speak, and setting some variables. Like, this is bullcrap, and I haven't seen any actual blind people on the Quorum list do anything with it. Like, they're all orgasming over the scene editor and "oh my gosh I could make audio games with this with so little code" type stuff, but I've not seen anyone use it. Luckily it doesn't seem that popular. Unluckily, Quorum was what I was introduced to in high school. Thankfully, my memory is crap so I easily forgot it enough to wire the bit of Python I do know now into my mind, and I hopefully didn't catch any bad habits from Quorum. But I'd hate to know what impressions blind people come away with from these classes. Also, it wasn't just the totally blind students who were taught Quorum, no, it was also the partially sighted students too. I fear that schools may just decide to go all in on Quorum, so that they have only one computer stuff thing to teach all their students. And that'd suck.
It's so easy to teach, too. They've got all the teaching material, all the basics (with oodles of spelling errors (hokeys. Like, seriously, who wrote this crap? It definitely wasn't a blind person.)), enough for teachers to have a good semester of material. Enough for the teacher to know just a tad more than the students, but once it's done, well, it's flushed out of the brain through earwax. Or something. And kids are all so impressed with this wizard who could hack into your computer in ten seconds and oh my actual gawid! It sucks. Like, imagine how much better the BeeWare project would be if *they* got that NSF money instead of Quorum? BeeWare is a GUI framework in Python that you can use to deploy your projects using native elements, so it's accessible. I mean, it's using a popular language that kids can pick up after high school, there are like at least fifty "learn Python" books that can surely be adapted for blind people to learn, and they can learn to use CLI tools like the Python REPL and Git and stuff they'd actually freaking use as a programmer. They would learn to interpret Python's error messages and deal with them.
I promise that if the place where I work ever gets a "learn to code" section for blind people, we're doing Python, or JS, or shoot even Ruby, over any toy.
Posts: 51 to 58 of 58
AudioGames.net Forum → Developers room → Is quorum a good place to learn how to coad
Generated in 0.045 seconds (19% PHP - 81% DB) with 9 queries