News for hackers. Stuff that matters.

Darkenetor

Re:Smaller Than Pixel Art: Sub-Pixel Art [video] on Thursday November 14, 2024 @12:08PM (#42138327)
There's a generator for those hidden subpixel message textures:

https://jsbin.com/korotaluso/edit?js,output

I have to recommend ComPressure for an accurate taste of the challenges in designing those kind of systems, on top of being my favorite Zachtronics-style puzzle.

Try the demo: https://store.steampowered.com/app/1528120/ComPressure/

Re:58 Bytes of CSS to look great nearly everywhere on Tuesday April 09, 2019 @04:56AM (#19612898)
Like this, unfortunately: https://gitlab.com/Darkenetor/userstyles-userscripts/tree/ma...

Firefox used to have an optional exception for icon fonts on the "allow custom fonts" switch which I found very sufficient. Not sure when it was removed or why, but that was a big loss.

Re:Show HN: Find recommendations from good people on Friday August 17, 2018 @03:24PM (#17785113)
Consider smile.amazon.com links or affiliates for charities like electronicfro-20 (EFF) though.
Re:Determinism in League of Legends: Unified Clock on Tuesday November 14, 2017 @02:15PM (#15698206)
Crashes and visual glitches are very rare, but there are a lot of bugs that just go unnoticed due to so much stuff happening at any time.

See [0] for the best example, but while now that those are documented it's clear that that's the worst case, a good number of champions contented for the Most Bugged crown over the years, even months after being released or still unfixed to date.

Other than that the new CEF client is still more buggy and unstable than the old one (same guy who made the above list also had one for the client although that hasn't been updated in months AFAIK) and at any time there has always been some sort of bug related to pathing, vision, or targeting.

Also I'm not 100% sure about the last year, but I think there as never been one since at least when I started playing in 2013 when some near-gamebreaking champion bug didn't make it to the "LTS" patch fixed for the LCS.

[0] https://boards.euw.leagueoflegends.com/en/c/bug-reports-eu/a...

One of my favorite games, MercurySteam's Jericho, is based on that mechanic. We definitely need more games to experiment with that.
At least on the beta channel that hasn't always been flawless. You can find many reports and patch notes regarding corrupted sessions and I've definitely lost some work over the years to that, the last time happening just half a year ago on the then latest version.
Reimplementing whole classes wouldn't have been a problem, just replace the original ones with a proxy to a copy.

Luckily console.log calls are actually excluded from the callers stack, so the only way is to change console itself. Just as evil if you also intercept console methods logging and return the original ones, but less of a mess at least.

Alas, for JS function.caller on a getter isn't standard but will work everywhere and should do it.
Re:Damn Small Linux on Saturday July 29, 2017 @09:00PM (#14883835)
Another still actively developed alternative is SliTaz [0].

Comes in images from 45Mb for a working desktop down to a bootable floppy, rolling release, runs in RAM with no persistence by default, and uses a custom package manager that, outside of having its own decently sized repo, can convert packages from Debian, RPM, Slax, Puppy, Slackware, NuTyX, Arch, Alpine, OpenWrt, 0Linux, paldo, Void and Tinycore according to the manual.

[0] http://www.slitaz.org/en/about/

The only solution is to always go for the HTTPS resource disregarding any suggestion. On browsers a strict configuration of Smart HTTPS [0] covers that, for everything else I think the best solution would be to intercept all HTTP traffic, request the HTTPS counterpart (and decide if falling back on failure is acceptable instead of just dropping the connection), then serving locally the decrypted response. Worse than properly requesting the right one from the start but harder enough to exploit.

[0] https://mybrowseraddon.com/smart-https.html

Re:The Rise of Pirate Libraries (2016) on Saturday July 22, 2017 @12:48PM (#14828028)
He read his father's without buying a second "license" himself for each book.
Re:How LZ4 works (2016) on Sunday June 25, 2017 @04:43PM (#14632059)
> lz4's decompressor is simultaneously simple and also the fastest thing aroud, at least at the moment

The proprietary LzTurbo [0] and free Lizard [1] both claim to be faster at decompression while having better compression ratios.

[0] https://github.com/powturbo/TurboBench

[1] https://github.com/inikep/lizard

Sounds like we have similar requisites. I settled on XMPlay (with an AIMP-like skin, way easier on the eyes), maybe it'll do?

https://support.xmplay.com/

https://support.xmplay.com/files_view.php?file_id=589

Re:Test-Driving a $200 Coding Font: Operator Mono on Wednesday November 09, 2016 @03:35AM (#12907877)
> Are there any other "non-usual" recommendations for writing code?

Pretty unusual, but after years I finally managed to find an alternative to Consolas I liked enough in Oxygen Mono—albeit I turned down a lot of really good fonts just because they had too much interline spacing and I couldn't tweak in Sublime.

Pretty much no Unicode support outside of common accents (good for every European language at least) and common programming symbols, though. Still I edited in em and en dashes and never found it lacking since.

Re:Neural Photo Editor on Saturday September 24, 2016 @09:28PM (#12573420)
Nnedi3 does.