It's especially important coming from someone like Terence Tao. If one of the best and brightest mathematicians out there can get a paper declined, then it can happen to literally anyone.
I also dislike the mindset that reading == intellectual. I know plenty of people who read a LOT but it's all Twilight or 50 Shades of Grey type stuff. And I know people who never read books, but are constantly watching documentaries or otherwise always learning.
People should be able to have hobbies and self improvement separate and each one delivered in whichever medium they happen to prefer.
(And I'm not trying to say bad things about either of the books I mentioned, but rather illustrate that certain types of books aren't going to make you a scholar, just because you read a lot. Read what you enjoy.)
As someone who reads a lot - probably more than 99% of people, I guess? - I mostly agree.
> People should be able to have hobbies and self improvement separate and each one delivered in whichever medium they happen to prefer.
Agreed.
> Read what you enjoy.
Semi-agree on this one.
There are different reasons to read. You can read for pure entertainment. You can read to learn things. You can read to be motivated. You can read to calm down. You can read to be part of the conversation.
Reading what you enjoy is good advice, and it's a valid approach, but so is reading something that's a little but hard and not very enjoyable, because you want to expand your knowledge, to learn and grow.
Read what you want for whatever reason you want, would be my recommendation.
> Read what you want for whatever reason you want, would be my recommendation.
My minor counter here is "don't let the reason you read [something] be because someone has made you feel like you should"?
I (personally) observe pushing through a thing you actively don't enjoy for external validation etc will just demotivate and bleed into the entire activity. That's not to say you should not try a book because somebody suggests/recommends it. But try, don't feel obliged in order to gain approval.
In the same vein, don't avoid reading "trash" for fun because people are asses. Be it young adult swashbuckling, adult romance or elves and wizards - don't let anyone make you feel bad about your preferences.
Thank you. this is much more what I was trying to convey by "read what you enjoy". I meant less "enjoy" as in "read because you enjoy it" and more "because you want to". If you WANT to read a book that challenges you, even if it's a bit tedious, then do so. But not because of the external validation.
if Shakespeare interests you, that’s fine. If you find him tedious, leave him. Shakespeare hasn’t yet written for you. The day will come when Shakespeare will be right for you and you will be worthy of Shakespeare, but in the meantime there’s no need to hurry things.
I don't know why this is, but every single documentary I have seen on a subject that I know something about has been factually atrocious. I strongly discourage people from watching any documentaries. Maybe a few nature-following-animals documentaries are OK, but for some reason I don't understand the average documentary is completely mendacious, with facts altered, twisted to be dramatic, omitted because they contradict some narrative, set pieces completely wrong, the creation of some narrative that doesn't even make real sense, etc.
It's not that there aren't awful books, of course there are, but documentaries seem to be almost uniformly awful, even ones very highly rated, where books are usually the actual source material anyway. On top of this, documentaries are much better suited as propaganda vehicles, and are often used that way.
Of course, it may be I am more sensitive than most people to these issues; other people may find lower truthiness levels acceptable if it's the only way to garner their attention.
Launching the Linux release and noticed in the logs:
Directories:User Directory: /home/bisby/Grayjay
And there is a directory there now. I absolutely hate having stuff automatically create anything in my home directory like this. Ideally, this should be following XDG directory guidelines on linux: https://specifications.freedesktop.org/basedir-spec/latest/
It's a "green m&ms" thing. If the developers can't be bothered to adhere to something as basic as XDG, they're getting a ton of other things wrong, and my life is too short to spend on buggy slop.
It is not really. If you show hidden files there is a ton of junk folders above your non hidden folders. along with more junk like folders generated by the system. I can't be arsed to go look up ways to keep them from returning or manage when I can just have my stuff reside in my own organization.
Windows user folder isn't better and Microsoft also adds tons of junk folders (3D Objects, Contacts, Links, Favorites, Searches, Saved Games)
Yes, it is. It takes only a few minutes to comply with the XDG spec. If an author can't be bothered to do that, he probably hasn't bothered to make his program secure, stable, or extensible either. XDG non-adherence is a strong negative quality signal.
Unless they simply are unaware, or hold a different opinion.
Frankly, I expect when something needs to create data without prompting the user, that data ends up in $HOME. I know where to look, and bonus points for an easy to manage folder with that data in it.
Grayjay dev here. If you want it to use your user directory like other apps, just remove the file called "Portable". Keep in mind that it just uses your working directory to write files otherwise.
Please just adhere to the XDG-standards. Although my co-poster here didn't use the most diplomatic way of phrasing their grievance Grayjay is better off if it sticks to well established standards.
You would probably look weird at an software that installs itself in C:\MYAWESOMEAPPLICATION instead of using the Windows program folder like literally every other piece of software (except for legacy stuff like LTSpice). Creating visible directories in the home folder without asking is the Linux equivalent of doing just that.
Check if the XDG environment variables are set and store your stuff in these places — as it is now can be used as a last resort fallback. For reading config/data you do the same.
Sorry, to be clear, I dont wan't Grayjay data in my user directory AT ALL. Portable is basically what I want, I'm just very untactfully dropping feedback about where the data is placed.
Even with the "Portable" file, it creates a directory `/home/bisby/Grayjay`. I don't want that. No app should ever put a file or directory directly in `/home/bisby` without me asking it to. The Linux standard for "where should an app put it's files" is defined the XDG spec that I had previously linked (https://specifications.freedesktop.org/basedir-spec/latest/).
The summary is that user specific data should live in $XDG_DATA_HOME and config should live in $XDG_CONFIG_HOME (and various other things like $XDG_CACHE_HOME). If these values are unset, there are predefined places to put the files (eg, data in $HOME/.local/share or config in $HOME/.config, cache in $HOME/.cache).
This puts all the Grayjay data in places like /home/bisby/.config/Grayjay (instead of /home/bisby/Grayjay) which is nested away inside a hidden directory and structured in a consistent way.
This would be the equivalent of putting data in %AppData% in windows instead of cluttering someone's "My Documents" (or whatever the modern equivalent of that is).
Some of the Linux decisions feel a bit like linux is a complete afterthought, but included because Linux users tend to agree with the FUTO philosophies. That is a reasonable thing given the Linux market share, and for "Build Version: 2" that I'm seeing the app info, I'm grateful that linux is included this early. This looks like it can probably replace freetube for me. However, it would go a long way if things are done to make sure they are done the "right way" on Linux (ie, on packaging and on directory specs).
Thanks for the work you've done on freeing up the web.
Well, ideally I would like the OS solving this problem by simply chrooting/sandboxing apps to their own little worlds, with a proper API giving them optionally a way to the user's file system, similarly to android and iOS.
That is possible on Linux [0], but this kind of separation comes with its own can of worms. However, if your only worry is access to folders, Flatpak applications keep all of their data in a folder away from your home directory and use "portals" to access your system [1]. The security of the sandbox is debatable [2], but I would say if your biggest goal is containing non-malicious but badly behaving applications from messing with your system, then it's a very good solution, given you are comfortable with using Flathub (as most distributions won't build Flatpaks) and with the performance/integration impact this distribution method has.
…and macOS. Sandboxed Mac apps get their own little home directory in `~/Library/Containers/`. To access anything else, they need to ask through system APIs.
If you're launching it as "Portable", and you're launching it from your home directory, it's going to place the mutable data in the current directory. This is very standard for portable apps.
So no, "portable" is not what you want. If you launch it as non-portable and it drops a folder in ~, then that is a problem.
In both modes it creates a ~/Grayjay directory, even when launching from ~/tmp/grayjay/Grayjay.Desktop-linux-x64-v2/ so ~/Grayjay was inevitable. In portable mode it makes the directory and does nothing with it. In non-portable mode it dumps a ton of data into the directory. I didn't pay attention to what the data actually was. So yes you're probably right.
But either way, Portable mode isn't behaving portably because it's touching directories outside of the current directory, and non-portable mode is putting data in ~/Grayjay instead of ~/.config/Grayjay so it doesn't do what I want it to do in any mode.
I'm quite happy actually that while this is a HUGE annoyance... It's also only an annoyance, and VERY simple to fix (as long as they do). Which means that this app is likely going to wind up as a daily driver for me once a few things get ironed out. I see the concept and structure of the app, and I like it.
>Parent is not wrong, but definitely could have some improved manners and tact.
I don't understand the weird tone policing that people are trying to do, there is literally nothing wrong with the parent's comment and pretending otherwise is weird.
I didn't see it as tone policing. From my point of view, I saw a very interesting application being shared and I hoped it would be good and prosperous so i can use it. When the first comment says "i hate that you do X" it is a bit discouraging towards a team of developers who have probably poured tons of free hours into making this. Words play with psychology, and it is my personal interest that these devs have good morale to make this app great, and that meand giving them feedback about obvious mistakes in a tone that does not hurt this morale. I hope that make sense
Comments like the one above do not belong here at all, it's childish, it brings absolutely 0 value to readers, and it's against the guidelines because it degrades the quality of HN for everyone.
What is considered impolite in the US or the UK is considered just being straightforward in e.g. Scandinavia.
I am German, we're kind of in the middle between someone from e.g. Finland and someone from e.g. or the UK or US with what we consider "ok" or rather crossing into rude territory.
A common exchange I witnessed in a meeting at work (Nokia):
Finnish developer: And if we follow this suggestion we will all look like idiots.
UK developer: I hear you.
Deciding which one is more impolite or impolite at all is left as an exercise to the reader. ;)
It's not normal to see someone doing x behavior, initiate a conversation with them and say "I hate x behavior." They're not tone policing you when they tell you that you come off like an asshole when you do that.
Actually, parent is wrong. You're not supposed to do that shit on Windows either. That's what AppData is for. Writing configuration files and folders to "Documents" or the user's home folder is sloppy shit.
I agree that this should be in the XDG directory or AppData, but be kind, y'all -- this is open source, it is a gift someone has labored over and given you. There are much nicer ways to suggest improvements than calling it "sloppy shit".
edit: it's not actually open source by the OSI definition it seems [1], but it is reasonably close.
Sure, things can always have gone better, but this is data loss/corruption territory. It's asking for trouble and hurt feelings. I think a strong response is ok.
I was very blunt and impersonal. People worked hard on a thing and my first reaction was criticism, without even the added overall view of "I love this thing, but here is a small thing that bothers me." I could have been more courteous and human about things.
I stand by the points I made, but I could have been friendlier. I normally make an effort to be friendly as I can about things, but I absolutely did not here. I hope that nothing I said came across as vitriol, but rather, valid criticism. I'm a strong believer in criticizing the things you love, but I need to remember that random comments on HN aren't the place where people know I love the thing, and my criticism needs context.
So no, it wasn't really that odious, but it was other things. Do I feel stricken with guilt or remorse about what I said? No. Could I have been friendlier? yes. Should I have been friendlier? Probably.
Even Windows has %appdata% which is where you put stuff on disk that you need to stash away. There's also function calls iirc which will give you a handle to a temporary file if you need it.
I use NixOS for the base system and Flatpak+Containers for everything else. I would rather keep it this way as it keeps everything nice and separated, and less chance of things breaking from Nix being rolling.
Grayjay dev here, we want to provide the app however people want to consume it (binaries, flatpak, appimage, ...) but it will take us some time to get everything as it should be.
Like others said, fighting the clean HOME fight is just draining and futile in the end. This script helps you identify low hanging fruit, though, where you can change their storage location with a simple envvar.
I have my home directory mounted via NFS, which has worked marvellously since gigabit ethernet became cheap 20 years ago but lazy/ignorant developers are making it less pleasant every year.
I've used xdg-ninja. And it does feel futile. Which is why my original comment comes across a bit more as frustrated/exasperated bug report than an HN comment.
Right now everything that xdg-ninja finds are all things with a .prefix and hidden. which is whatever. if this was ~/.grayjay, I probably would have rolled my eyes but not even bothered to comment. I'm not a stickler about XDG, but I am a bit of a stickler about not cluttering my home directory.
I prefer Signal, but a lot of my friends are iPhone users and just use iMessage amongst themselves. Android RCS to iMessage is a step up over SMS (RCS doesn't result in super low res videos and pictures).
The thing about social interactions is that you don't always get to dictate "We're using Signal now." And if the other people are using iPhones and aren't going to use Signal, RCS makes sense.
RCS does NOT work properly on custom roms unless you have some google integrity check hacks mixed in, and its a constant cat and mouse of google breaking those hacks... so if your messages are timely, you want SMS, or a google approved android.
Some days I don't shower and I look like a gremlin and I just want to get work done without worrying about my appearance. I don't work in a position where appearance actually matters for anything.
Also, what is the value in having it on? I've worked places where they want your cameras on during all staff meetings with 100+ people on them. I don't actually understand the upside to having it on.
If leadership wants to have their own cameras on because they feel it conveys intent and emotion better when I can see their face... that sounds like they made a decision. Leaders often have to lead with emotion so wanting to convey emotion better might be a good choice for them. Most of the meetings I'm in are about facts. And I dont feel like I need facial expressions to express data more effectively.
Anyone who worked pre-Zoom for a large corporation with offices spanning multiple cities will remember having conference calls simply because everyone was in office, but there were 2-3 offices to connect together. And you could see people in your own office, but you couldn't see people in other offices, and not being able to see people was never an issue back then. This isn't even a new concept. It worked for DECADES just fine.
It wasn't just fine, otherwise there would be no need for business travel, and Skype's been around since 2003 so video calls have also been around for decades at this point.
I haven't found Copilot that useful, except as a fancy autocomplete. However, recently I've had Claude solve a lot of problems for me, including some with fairly tricky math. Nothing that I couldn't solve myself. But definitely things that would have taken me from 30 minutes to a few hours, and which with Claude helping me were solved in 5-10 minutes.
As an example from yesterday, I had written a function that returned the position of the camera in 3d space relative to an object. I'd written it fairly quickly and naively using just the direction vector and it mostly worked, but then I realized when the camera was overhead facing down, I also needed to consider the orientation of the camera. I fed my function into Claude, explained the problem without my hunch on how to fix it because it's always interesting to see if I'll get a different solution. But Claude was able to get the same solution as me, and do the maths required, in one go.
I generally don't try to get answers for things I don't already mostly understand anyway (sounds like you did this as well), because if what I get back is a hallucination, I don't want to spend added time figuring that out, and then I'm back at square one and have lost time.
> a popular new game by Valve is properly newsworthy to report upon
If you've been explicitly asked to not report on a thing, and then you do. That's a jerk move. There shouldn't have to be legal incentive to do the right thing.
> a big enough company to be able to require and enforce binding NDAs and they chose not to in this instance
Valve isn't Nintendo. Their goal isn't to lawsuit their fans into submission. The splash screen saying "Don't share stuff" that the Verge clicked "OK" on is not necessarily a legally binding NDA, but it was clear what was being asked, and the Verge broke that contract (even if it was just a social contract). And Valve enforced the contract by banning them. No legal repercussions, just "you didn't hold up your end of the deal, so we're removing you from the deal" (and now Valve knows for the future that the Verge and this author in particular are the type to not follow simple instructions).
Access to the game is pretty readily available. If you want in, you can get in. People playing have to agree to the "Don't share stuff, this is just alpha" screen, and they know they are playing alpha software. People reading (or more likely, skimming) an article, don't have to agree to "I realize this isn't final software." So while nothing is so secret that it needs legal enforcement, different circumstances of how you discover something may affect your view on it.
> If you've been explicitly asked to not report on a thing, and then you do. That's a jerk move. There shouldn't have to be legal incentive to do the right thing.
NDAs are the way companies ask a reporter not to report on a thing. It's a valid agreement between two parties. The jerk move is expecting people to adhere to your version of what "the right thing" is. Not everybody shares your opinion on the matter, hence the existence of NDAs.
NDAs are the way companies DEMAND a reporter not to report on a thing. Again, Valve is not trying to turn this into a legal mess. There were no repercussions to the Verge other than losing access to the beta. I don't think that companies being rigid soulless entities is a good thing. If this is the kind of thing that makes Valve go "I guess we need to invite fewer people and make everyone sign an NDA" then the world is robbed of just a bit of fun. And that makes the Verge's actions a jerk move at an even deeper level.
The actual prompt that shows up when you launch the game says:
> Early Development Build
> Deadlock is still early in development, with a lot of temporary art and experimental gameplay. Do not share anything about the game with anyone.
> [ OK ]
They make it pretty clear what they want in exchange for playing the game.
> Expecting people to adhere to your version of what the right thing is.
You think that asking people to do something as a human, instead of sterilizing the interaction and bringing lawyers into it... that's the jerk move.
There isn't much wiggle room on what is right here. If the Verge didn't want to adhere to the agreement, they could have just closed the game. So they agreed to not share the information, and then did anyway. That's dishonest, and a jerk move. They weren't being dishonest to expose a larger crime (this wasn't whistleblowing), they were just dishonest for their own benefit.
I would hope you're not the type of person who shares other people's secrets, because you're not legally obligated to since you never signed an NDA.
> with particularly dangerous temperatures forecasted for cities such as Chicago, New York, and Boston
> A heat dome that descended on Portland, Oregon,
> FastCompany is an American company
American companies tend to write for American audiences, especially when the article leads off with a bunch of American anecdotes.
If it was fastcompany.co.uk and it started off writing about how dangerous temperatures were going to be in London, and Liverpool, I would have assumed it was British.
The "intended audience" isn't just "people who can speak this language", and there are plenty of other context clues other than just "it's in English"
