tamas

@tamas@hachyderm.io

C++ enabler at NetworkOptix. Formerly Plex.

This profile is from a federated server and may be incomplete. Browse more on the original instance.

tamas, to GNOME

The performance improvement in the #GNOME Files app is fantastic (particularly, opening a directory with many entries is almost instant now). Thanks to whoever fixed that 🙏

fasterthanlime, to random
@fasterthanlime@hachyderm.io avatar

mhhhhhhhHHHHH.

tamas,

@fasterthanlime what's up with the kerning?

seanmonstar, to rust
@seanmonstar@masto.ai avatar

curl asks: "hyper, is it worth it?"

A fair question. It costs both projects to maintain the support.

Generally, none of my "users" ask for it, and my lack of experience in C and a foreign code base means I can't push on it.

I do think it would be beneficial for the Internet. But after some years of pointing out the "few" pieces left, no one has taken to seeing through.

It needs someone. #rustlang

https://mastodon.social/@bagder/112279827007945931

tamas,

@seanmonstar Is the missing features the actual problem or the maintenance cost? Wouldn't essentially the same problem remain if someone stepped in and implemented the missing pieces?

DenisCOVIDinfoguy, to metallica
@DenisCOVIDinfoguy@aus.social avatar

Court Rejects Metallica’s Insurance Lawsuit Over COVID-Cancelled Concerts | billboard.com

"A California appeals court has rejected Metallica’s lawsuit demanding that its insurance company pay for more than $3 million in losses stemming from concerts that were canceled due to the COVID-19 pandemic."

#Metallica #COVIDcancelledconcerts #PandemicIsNotOver @auscovid19

Source: https://www.billboard.com/business/legal/taylor-swift-metallica-court-rejects-lawsuit-covid-cancelled-concerts-1235636969/

The decision, issued Monday (March 18) by California’s Court of Appeal, said that six COVID-cancelled 2020 shows in South America were not covered by Metallica’s insurance policy with Lloyd’s of London, thanks to a clear exclusion in the contract for any losses stemming from “communicable diseases.” The legendary rock band had argued the case should have gone to trial, since a jury could have decided that non-COVID reasons led to the cancellations. But Justice Maria Stratton, improbably citing Swift, said it was “absurd to think that government closures were not the result of Covid-19.” “To paraphrase Taylor Swift: ‘We were there. We remember it all too well,’” the justice wrote. “There was no vaccine against Covid-19 in March 2020 and no drugs to treat it. Ventilators were in short supply. N-95 masks were all but non-existent. Patients were being treated in tents in hospital parking lots. The mortality rate of Covid-19 was unknown, but to give just one example of the potential fatality rate, by late March, 2020, New York City was using refrigerated trucks as temporary morgues. People were terrified.” Metallica’s case is one of many that have been filed by musicians, venues, bars and other businesses seeking insurance coverage for harm caused by the outbreak of COVID-19, which led to months of severe travel restrictions, forced closures and bans on large gatherings. .
Court documents show that in May 2020, the band submitted a loss of $3,234,569 stemming from the cancelled shows, covering things like $184,996 in payroll for retained crew members. But citing the disease exclusion, the insurer quickly denied the claim: “Unfortunately we have to advise that no coverage is afforded for this matter under this Policy,” the company wrote in a June 2020 response letter. In December 2022, a Los Angeles judge rejected Metallica’s case and the various arguments for why Lloyds should have paid for the concerts — including ruling that the cancellations were caused by travel restrictions that were “a direct response to the burgeoning COVID-19 pandemic.” Appealing that decision, Metallica argued that a jury might have found a different cause for the concert cancellations. The band’s attorneys pointed to the fact that venues later reopened and the shows were performed in 2022, “despite the ongoing presence of COVID.” The judge also rejected various other arguments from Metallica, like the claim that the policy did not cover COVID cancellations because it did not specifically use the term “virus”: “The insurance policy definition of communicable disease does not refer to any pathogens nor does it limit the exclusion to only those communicable diseases caused by specific pathogens.” Attorneys for both sides did not immediately return requests for comment.

tamas,

@Tooden @DenisCOVIDinfoguy @auscovid19 Honestly, as much as I despise this, the fact that insurance companies won’t pay even when it’s a once in a lifetime event (i.e. the customer’s bet on an unlikely disaster came through) is maddening. EVEN when the customer has expensive lawyers.

fasterthanlime, to random
@fasterthanlime@hachyderm.io avatar

every week I ask the internet "how do I watch show X legally?" and every week the internet looks back at me and laughs

tamas,

@fasterthanlime Try watch.plex.tv - it tells you where a certain show or movie is available at the moment.

tamas, to GNOME

It's really difficult to find a custom #gnome shell theme that is not a (usually poor) remake of macOS or Windows looks :blobfoxsad:

tamas,

@cassidy I decided to switch to light themes everywhere and many parts of the shell remain dark which looks out of place for me. Most prominently, the top bar and the dock, as well as the "status menu" (I don't know what it's called, I mean the menu in the top right corner that brings up the volume control and others).

You could say it's an accessibility need because I realized that light backgrounds work better with my astigmatism. But I'm not saying I'm suffering because of the dark colors.

tamas,

@cassidy So, to be clear, the reason I was looking a custom theme was an aesthetic choice/preference for a uniform look; even though the motivation to switch to light mode is basically accessibility.

hazelweakly, to random
@hazelweakly@hachyderm.io avatar

I have a new blog post for y'all on observations of leadership. I've been in a really interesting position lately: I started off as an IC in my current job, stepped into an interim director + manager role, became the manager once we hired a director, and finally stepped into a principal architect role in Q1 of this year.

Which is a lot! And I've been wanting to write up on what I learned, but haven't quite had either the time or the inspiration until recently

tamas,

@hazelweakly What does IC mean?

tamas,

@hazelweakly Gotcha, thanks!

fasterthanlime, to random
@fasterthanlime@hachyderm.io avatar

bearbee alert

https://neal.fun/infinite-craft/

tamas,

@fasterthanlime I want to stop

jmb, to Magic
@jmb@mastodon.me.uk avatar

For my #magic ( #conjuring ) hobby, I catalog videos that make up online courses (most only one video, some multiple) by writing a “timeline” for each, signposting for people where to find particular items in each video. This is currently just written up in Apple Pages, exported as a PDF and shared by Dropbox link. I started a website a while ago using @django to make it more accessible and searchable but never really came up with a completely satisfactory data structure (see pic)

tamas,

@jmb Wagtail is a CMS that you can customize and program around. If you already have a Django site, you might end up fighting Wagtail more than what you save on existing structures. I’m developing a wagtail site, but I started from scratch and I only have a few special pages, the rest is CMS content.

tamas,

@jmb Well then, maybe it's worth considering!

bluejekyll, to rust

I follow # C mostly as a means of practicing patience so that I break my conditioning of responding immediately to quick takes of “C is great!” It takes all my power to avoid pointing to all the security issues, bugs, and basic errors that languages like #rust prevent.

Which brings me to a question, when someone posts to a # you follow, it seems like it’s probably rude to respond (unless it’s an obvious question), but why did they use a # in the first place if not for randos to comment?

tamas,

@bluejekyll It’s one thing to chime in on tags you follow (which is totally fine), but another to chime in with the most stereotypical / annoying “Rust take” when security or the classes of bugs that Rust prevents weren’t the topic. Imagine someone coming to your threads just to tell you about how Rust safety is a lie because the unsafe keyword exists or because the Rust compiler is built on top of LLVM, a C++ project.

markwalker, to django
@markwalker@fosstodon.org avatar

Does anyone know of a reputable source for an article on using python -m

I recommend it to people but lack a credible source to back it up. Just had someone say that it's "not recommend or the convention. There are no benefits."

tamas,

@markwalker Can't give you an article, but I can tell you my reasoning for using it: with the mess of different ways of installing Python software, it's easy to end up in a situation where the tool you are calling by name is actually running a different interpreter than what python comes first in your PATH. This is particularly problematic for pip because it might install a given package for a completely different python installation without any errors.

sudoaptgetlife, to plex
@sudoaptgetlife@mastodon.social avatar

Has anyone else noticed a problem with #plex subtitles on ad supported content? I was watching a foreign language film with subtitles and every time it went to an ad break Plex would disable the subtitles and the only way to get them back was to go to playback settings and turn the subtitles off and on each time. This is extremely annoying to have to do this, considering how many ad breaks they insert into the average film on this platform.

#film #movies #cinema #streaming @plex

tamas,

@sudoaptgetlife @plex Plex does not decide which ads are shown. It sends a request to the ad network, gets an ad, plays it and the reports back to network (that the ad was played). Also @plex doesn't seem like an official account.

thomasfuchs, to random
@thomasfuchs@hachyderm.io avatar

The only reason big tech companies want you to believe in the LLM hype is that they tend to have massive datasets they can mine and that hype creates a path for them to sell this data for a lot of money.

Think about all the messages and conversations on Facebook. Or comments on YouTube. Or code on Github. Or posts on Reddit. Or your tweets. Etc. Etc.

LLMs are a shit technology that has almost no actual use to do anything good, other than make corporations and their shareholders richer.

tamas, (edited )

@thomasfuchs Eh, in general I agree, but Github Copilot is really useful (not the chat, only the completion engine). That is setting aside the ethical concerns, and I’m only talking about the technological usefulness.

tamas,

@thomasfuchs That a different problem (and I agree, it’s a net negative for learning). But your initial claim was that LLMs are not actually useful. My counter-argument to that is Github Copilot, which I think is really useful. There are many examples of not actually useful LLMs, of course.

thomasfuchs, to random
@thomasfuchs@hachyderm.io avatar

If you use Substack, move to another platform.

If you subscribe to something on it, pressure the authors to move to another platform. If they don’t, cancel your subscription.

Substack refuses to stop financing Nazis.

tamas,

@thomasfuchs @sbi Nilay Patel’s interview with the CEO is a sight to behold. Apparently he didn’t change his stance, even though he was very embarrased https://www.theverge.com/2023/4/24/23696530/substack-notes-moderation-bigotry-chris-best

bagder, to random
@bagder@mastodon.social avatar

Please don't make this a new trend. 😕

(issue closed by bot because the user filing the issue has not starred the repository...)

tamas,

@bagder Where is this?

tamas,

@popey @bagder Embarrassing it was added in the first place, but at least they shut it down.

syntaxseed, to webdev
@syntaxseed@phpc.social avatar

So... #HTMX is cool, but not quite what I was hoping for. Great for replacing messy AJAX calls but what if some of your data is generated client-side?

I was hoping I could hook the HTMX triggers & DOM replacement features with a mix of server calls or client side JS.

It's tough to try for a SPA replacement or even just a robust app if you shove too much work through server calls & users on slow networks get neglected.

This doesn't really replace front end frameworks.

#WebDev

tamas,

@syntaxseed Check out AlpineJS if you haven’t already. For me, it strikes a very good balance (I’m not a web developer and large frameworks and project setup is really overwhelming)

tamas, to ubuntu

Does have special patches for fractional scaling under that is lacking? I didn't see the option with Fedora. Not keen on switching to KDE. In any case, it's really impressive that Asahi & derived distros are at this level already.

AstraKernel, to rust

✨ #RustQuiz

What will be the output?

You can vote below 👇

#rustlang #rust

// Y: struct Student { age: u8, score: u8, name: String, } fn main() { let s = Student { age: 22, score: 98, name: "Hello".to_string(), }; match s { Student { age, score: 98, name:_ } => { println!("{age}"); display_details(&s); } _ => println!("NOP"), } } fn display_details(s: &Student) { println!("{}", s.name); }

tamas,

@AstraKernel This is really interesting, but I think it will be a compilation error because of the &s in both cases. But if I'm wrong, that's a neat thing I haven't thought of before in match.

tamas,

@AstraKernel Surprising result, it looks like other people were confused as well. Do you have an explanation for this behavior?

  • All
  • Subscribed
  • Moderated
  • Favorites
  • •
  • JUstTest
  • mdbf
  • ngwrru68w68
  • tester
  • magazineikmin
  • thenastyranch
  • rosin
  • khanakhh
  • InstantRegret
  • Youngstown
  • slotface
  • Durango
  • kavyap
  • DreamBathrooms
  • megavids
  • tacticalgear
  • osvaldo12
  • normalnudes
  • cubers
  • cisconetworking
  • everett
  • GTA5RPClips
  • ethstaker
  • Leos
  • provamag3
  • anitta
  • modclub
  • lostlight
  • All magazines
    •