|
ultrafilter posted:Designers don't get to unilaterally decide whether there's a login required to interact with a system. Foxfire_ posted:"Can we change our online-only game to not require an account?" is not a useful question for a UI designer to be asking or making mockups for, especially at the point where you are hashing out the details of a screen. it's a useful assignment, folks are projecting onto it and just going hog wild
|
# ? Sep 21, 2022 19:26 |
|
|
# ? May 28, 2024 14:44 |
|
a lovely web game for a class? wow, how are you going to ship that without a *checks notes* extensive EULA flappy bird's EULA was one for the ages, truely a classic in the genre, that's why it made such a splash right?
|
# ? Sep 21, 2022 19:28 |
|
Who should be deciding what the user-facing authentication flow is if not designers?
|
# ? Sep 21, 2022 19:29 |
|
Mullvad's is still the greatest onboarding process ever (for a paid service): https://mullvad.net/en/account/#/create/
|
# ? Sep 21, 2022 19:41 |
|
This depends a lot on how the task was presented. Given an example that includes a login, I'm not surprised that most students make the implicit assumption that it's a requirement. The best answer might actually be to request more information.
|
# ? Sep 21, 2022 19:43 |
|
JawnV6 posted:you've never been on a project early enough???? wow, alright, that's okay, but it's hardly something to brag about in a scenario like this. you could learn something. Assuming that you're able to just change a project that has already started and been brought into is certainly a possibility. Alternate possibility: "uh, no, we kind of need this actually since people need to log in to play World Of Dumb Slapfights or whatever we're calling it this week. Can you please just make it not idiot programmer art and a weird workflow?" I get it, design is important, but students also need to make assumptions like "the instructor asked us to work on this so I guess we should work on it?" if they want to keep the GPA their scholarship is riding on. Maybe there was a sense of ~~~OwNeRsHiP~~~ present in the assignment but that definitely wasn't communicated. Especially since we're just coming off the "the junior dev threw away a perfectly good and useful piece of equipment for bespoke garbage that's broken and now I have to deal with it" chat. Volmarias fucked around with this message at 20:06 on Sep 21, 2022 |
# ? Sep 21, 2022 20:03 |
|
Someone taking a UI class presumably is learning to be entry level to start with, they're not necessarily going to be expected to ask big brain questions about the very premise of the product they'll be working on, although sometimes that would be useful - however, focusing on actually dealing with the task at hand instead of second-guessing it is also a useful skill, and is probably more consistent with the vast majority of work of even higher level UI designers. But regardless of all that, I want to iterate that if I wanted my students to ask these questions, and 100% of the time they ended up not doing that, I would, with respect, question the curriculum leading up to that point, or the formulation of the problem I've given them, before making fun of them. e: fb
|
# ? Sep 21, 2022 20:22 |
|
see, that's more along the lines of the silly quibble I wanted to throw in here, I'd call that level of changing expectations UX instead of UI, then the nerd brigade strolled in assuming they'd baked a super cool login system that's getting chucked out by the
|
# ? Sep 21, 2022 21:09 |
|
I see and know the value of onboarding new users smoothly, but calling it the nerd brigade isn't helping here. This sounds more like someone has studied a UX flow on mobile devices or Steam where there is already an identity system in place and, seeing no login, just assumed everything was doable without a login. It's a generalized question with a gotcha answer based in no concrete reality or expectation. Ranzear fucked around with this message at 22:07 on Sep 21, 2022 |
# ? Sep 21, 2022 21:26 |
|
learning that UI goes beyond polishing corners is fine, actually appreciating disciplines outside of coding is fine, actually being a nerd in the nerd brigade is fine, actually
|
# ? Sep 21, 2022 21:50 |
|
JawnV6 posted:learning that UI goes beyond polishing corners is fine, actually Responding to what people actually said is even better!
|
# ? Sep 21, 2022 21:57 |
|
if 100% of students get a question wrong, it means they weren't taught the subject matter they were being tested on. simple as that.
|
# ? Sep 21, 2022 22:01 |
|
Absurd Alhazred posted:Responding to what people actually said is even better! i agree, you're the one who injected the word "wrong" to describe the assignment when the UI teacher never used that word and it derailed the entire conversation. would you care to walk that one back and explain to folks like cheetah here why reading what folks actually said is even better?
|
# ? Sep 21, 2022 22:03 |
|
Volte posted:Who should be deciding what the user-facing authentication flow is if not designers? Design, product and engineering should all have inputs into how it works.
|
# ? Sep 21, 2022 22:13 |
|
JawnV6 posted:i agree, you're the one who injected the word "wrong" to describe the assignment when the UI teacher never used that word and it derailed the entire conversation. would you care to walk that one back and explain to folks like cheetah here why reading what folks actually said is even better? No, because this is what I was responding to (Bolding mine): lord funk posted::nods: Please explain how this response (bolding mine): Absurd Alhazred posted:I would also suggest that if 100% of your students get the answer to your question wrong, it might be a problem with the curriculum leading up to the question, or the question itself, not the students. is an "injection" of the word "wrong". My position is that this is a legitimate restatement of what OP said, and didn't derail anything.
|
# ? Sep 21, 2022 22:17 |
|
Absurd Alhazred posted:My position is that this is a legitimate restatement of what OP said, and didn't derail anything. well if you're this hepped up on bad faith there's little i can do to assist things, but here we go the fact that there's a better answer available doesn't make the submitted assignments "wrong" or earn them a failing grade, as we can see your callous and careless assignation of "wrong" caused folks to assume that was the case. it's an entirely legit pedagogical approach to getting UI/UX folks to take that step back and realize that their domain is broader than taking whatever engineering chucked over the fence and said implement. im also taking lord funk to be a good educator, and isn't gleefully handing out failing grades to 100% of the students semester after semester and this "task" (note: not assignment!) is serving the goals of the course. but you could be right, he could be doing the same thing repeatedly and weeping when nobody figures it out, marking down students on a gotcha question and regaling us with the story about how they're a lovely teacher with zero pattern recognition. personally, that second one seems more of an affront than anything ive said here, but you do you
|
# ? Sep 21, 2022 22:26 |
|
in either case (poor teaching or teaching method where you expect most people to not figure it out), coming onto the internet to shake your head ruefully about how nobody did it right feels like a bad thing to do (and is fairly inconsistent with the idea that they expected people to not figure out what they wanted)
|
# ? Sep 21, 2022 22:29 |
|
JawnV6 posted:well if you're this hepped up on bad faith there's little i can do to assist things, but here we go quote:the fact that there's a better answer available doesn't make the submitted assignments "wrong" or earn them a failing grade, as we can see your callous and careless assignation of "wrong" caused folks to assume that was the case. it's an entirely legit pedagogical approach to getting UI/UX folks to take that step back and realize that their domain is broader than taking whatever engineering chucked over the fence and said implement. quote:im also taking lord funk to be a good educator, and isn't gleefully handing out failing grades to 100% of the students semester after semester and this "task" (note: not assignment!) is serving the goals of the course. but you could be right, he could be doing the same thing repeatedly and weeping when nobody figures it out, marking down students on a gotcha question and regaling us with the story about how they're a lovely teacher with zero pattern recognition. personally, that second one seems more of an affront than anything ive said here, but you do you
|
# ? Sep 21, 2022 22:35 |
|
oh yeah breaking quotes up and inlining a bunch of poo poo, hallmark of good faith debate that it is,
|
# ? Sep 22, 2022 01:12 |
|
JawnV6 posted:oh yeah breaking quotes up and inlining a bunch of poo poo, hallmark of good faith debate that it is, You have my permission to stop posting and just assume that you've won the debate, if it will end this tedious derail.
|
# ? Sep 22, 2022 01:21 |
|
I have two thoughts on this. One is that "improve this basic UI design" is a prompt without enough context to meaningfully do anything other than turd-polishing -- you cannot reasonably assume that the "correct" solution to the prompt is "get rid of the login system" if the original design has a login system, because a trick answer like "the login system is now the problem of the launcher team instead of the game client team" is a much more complicated thing to do for everyone involved and really isn't something you want to do as a sweeping change -- you plan that from the start. The other is that the idea of "no menus, jump straight into the gameplay" is one of the more annoying bits of modern game dev UX fuckery because every game on the planet has incredibly loud sound defaults and I don't want to sit through a whole cutscene and tutorial zone with my headphones on my shoulders to prevent every loud noise from perforating my eardrums. Launch game for the first time, fiddle with settings, then start playing.
|
# ? Sep 22, 2022 02:14 |
|
I do really like the idea but wonder if it would be better as a question asked in lecture with a humorous response when students inevitably fail it, instead of as part of a graded assignment
|
# ? Sep 22, 2022 02:15 |
|
end of the day, the "let's get rid of the login screen" example started with not enough info to determine whether getting rid of the login screen is even feasible. this bullshit is frankly making me sad
|
# ? Sep 22, 2022 03:26 |
|
On the front end, someone in UI/UX might have a brilliant idea that gets rid of the login screen, but they have to collaborate with the backend people who care about infosec vulnerabilities and poo poo. This conversation has turned loving ridiculous. But it's mostly JawnV6's fault, because as much as he wants to pretend he has the most open mind, he's taking everything everyone else says in the worst faith possible. Jesus
|
# ? Sep 22, 2022 04:09 |
|
If none of the students arrive at the answer the teacher expects, then that is the teacher's failure, not the students'.
|
# ? Sep 22, 2022 05:46 |
|
DaTroof posted:end of the day, the "let's get rid of the login screen" example started with not enough info to determine whether getting rid of the login screen is even feasible. this bullshit is frankly making me sad I've heard in a youtube video, which is my extend of game design knowledge, that time to play is important and should also be short. Maybe this is the concept the students were meant to ... pick up on their own? I don't know it's all conjecture. but crucially: QuarkJets posted:If none of the students arrive at the answer the teacher expects, then that is the teacher's failure, not the students'. my personal horror of today: Entity Framework Core creating migrations with nothing in them, and when run in verbose mode tells me nothing about why the migration is empty. I hate EF so so much. champagne posting fucked around with this message at 09:31 on Sep 22, 2022 |
# ? Sep 22, 2022 07:46 |
|
My experience with teaching undergrads is that if you want them to go with answers like "let me just cut out these login screens" on their own, you need to go absurdly over the top. Have a screen that asks them to create a local account. Then pop up a screen that forces them to create a publisher account. Then pop up a screen for creating an online play account. Then make them select a name for the character. And so on Or make it super duper clear that the game is single player, offline only, so online accounts are stupid.
|
# ? Sep 22, 2022 09:15 |
|
Xarn posted:My experience with teaching undergrads is that if you want them to go with answers like "let me just cut out these login screens" on their own, you need to go absurdly over the top. Have a screen that asks them to create a local account. Then pop up a screen that forces them to create a publisher account. Then pop up a screen for creating an online play account. Then make them select a name for the character. And so on If this is your learning goal just put it on a slide, then explain why it's on there.
|
# ? Sep 22, 2022 09:32 |
|
Spent half the afternoon tracking down a failure in a new deployment of an app (an engineering tool, not a customer-facing app) to a new version of a JS dependency. Well, should be no problem, right? Just stop it from pulling in the new version. The person who originally created the app intended its dependencies to be maintained (or whatever is the term) using "yarn". They apparently took steps to pin dependency versions in the way "yarn" wants you to do things. Then someone later came along and added an automated deployment process using "npm" and something called "react-scripts". This thing is pulling in the latest version of the dependency despite the fact that there's an entry in "packages.json" telling it to use a specific version. Ok, internet says I need to take "package-lock.json", amend the version of the dependency and change the name to "npm-shrinkwrap.json", and it will use that exact version. Nope, somewhere along the way the deployment process overwrites "npm-shrinkwrap.json" and carries on using the latest version of the dependency. Look all over to figure out what it's doing, all I can find is that it is running "npm install", which Stack Overflow assures me respects "npm-shrinkwrap.json". There's a command in npm, "npm explain", which is supposed to tell you why it's pulling in a particular version of a dependency, but it's only available in a later version than what's installed on the build server. and considering how loving flaky all these tools are I have no doubt that if I try to update the version on the build server I'll break everything for all the teams using it and get half a dozen people mad at me. I do .NET languages, I shouldn't have to mess around with this stupid Javascript horseshit in the first place. Feeling about ready to do violence here.
|
# ? Sep 22, 2022 19:06 |
|
Hammerite posted:Spent half the afternoon tracking down a failure in a new deployment of an app (an engineering tool, not a customer-facing app) to a new version of a JS dependency. Well, should be no problem, right? Just stop it from pulling in the new version. it's job security for web developers. i've seen enough projects that only work if they use the one specific version of node.js that doesn't crash if you try to use a certain minifier etc. it's pretty dumb. granted everyone who works in .net (at least back in my days as a .net developer) has googled something along the lines of "how do I get projects that use different versions of newtonsoft json to work together?" and learned a whole bunch of bullshit about binding redirects.
|
# ? Sep 22, 2022 20:52 |
|
This may not be a complete solution because it sounds like fucky things are going on, but you want to use npm ci instead of npm install. npm install does completely insane things, while npm ci does a much more straightforward "install the versions the package says to install".
|
# ? Sep 22, 2022 21:35 |
|
Bruegels Fuckbooks posted:granted everyone who works in .net (at least back in my days as a .net developer) has googled something along the lines of "how do I get projects that use different versions of newtonsoft json to work together?" and learned a whole bunch of bullshit about binding redirects. Microsoft adding JSON support to the BCL has been an absolute godsend for this, and I don't even mind that they're taking their time to get to feature parity so they can do it right. Now, let me just make sure that there's nothing external blocking me from upgrading this project to a framework that includes System.Text.Json...oh...oh no
|
# ? Sep 22, 2022 21:46 |
|
Hammerite posted:The person who originally created the app intended its dependencies to be maintained (or whatever is the term) using "yarn". They apparently took steps to pin dependency versions in the way "yarn" wants you to do things. Then someone later came along and added an automated deployment process using "npm" and something called "react-scripts". Okay, this is a major problem right here. I'm not trying to defend the hosed-up JavaScript ecosystem, but even if you're stuck in that hosed-up ecosystem, you can't just switch between npm and yarn all willy nilly and poo poo. Somebody hosed up
|
# ? Sep 23, 2022 03:53 |
|
raminasi posted:Microsoft adding JSON support to the BCL has been an absolute godsend for this, and I don't even mind that they're taking their time to get to feature parity so they can do it right. Now, let me just make sure that there's nothing external blocking me from upgrading this project to a framework that includes System.Text.Json...oh...oh no More like "make sure you're not de/serializing anything but the simplest, most primitive POCOs". One of Newtonsoft.Json's strengths is that it has a veritable Swiss Army Knife of option to control your serialization process, at pretty much any level of granularity you want - attributes, converters, contracts, you name it - and I don't think STJ will ever achieve 1:1 compatibility. However, for greenfield stuff, STJ all the way.
|
# ? Sep 23, 2022 08:40 |
|
I put it back to using Yarn and it's working now. Rage has subsided, the solemn intent to destroy all JavaScript has not.
|
# ? Sep 23, 2022 11:10 |
|
NihilCredo posted:More like "make sure you're not de/serializing anything but the simplest, most primitive POCOs".
|
# ? Sep 23, 2022 15:12 |
|
SupSuper posted:Swiss Army Knife serializers seem to be a pretty good recipe for security holes. On the other hand. having to wrote one more line of code than theoretically necessary to parse json is a tragedy, so
|
# ? Sep 23, 2022 15:22 |
|
Hammerite posted:I put it back to using Yarn and it's working now. Rage has subsided, the solemn intent to destroy all JavaScript has not. I'm going to get a "gently caress javascript" tattoo somewhere just to make the interview process easier.
|
# ? Sep 23, 2022 15:56 |
|
SupSuper posted:Swiss Army Knife serializers seem to be a pretty good recipe for security holes. Isn't it probably better than writing your own? Like presumably the big libraries are going to be more familiar with the edge cases/etc. They're certainly bigger targets but like, I don't trust myself to know JSON deserialization security better than any of the big libraries.
|
# ? Sep 23, 2022 18:16 |
|
|
# ? May 28, 2024 14:44 |
|
Jabor posted:You have my permission to stop posting im sorry this thread can't imagine a question that doesn't have a binary RIGHT/WRONG answer especially in a UI space, it's a severe limitation to carry out there as someone working with other humans and i sincerely hope this is just trolling me at this point
|
# ? Sep 23, 2022 20:27 |