let's talk about Bridgerton tea, my ask is open
No title available
#extradirty
tumblr dot com
will byers stan first human second
JVL
wallacepolsom
No title available
dirt enthusiast
🪼
blake kathryn
PR's Tumblrdome
noise dept.
"I'm Dorothy Gale from Kansas"
roma★
Janaina Medeiros
taylor price
Product Placement
Cosmic Funnies
AnasAbdin
seen from Malaysia
seen from North Macedonia
seen from Vietnam
seen from United States
seen from United States
seen from United States
seen from Canada
seen from China
seen from United States
seen from United States
seen from Indonesia
seen from United States
seen from Brazil
seen from United States
seen from United States
seen from United States
seen from United States
seen from United States
seen from United States
seen from United States
@frog707
NB 30
Today binaries were finally published for version 30 of Apache's NetBeans integrated (software) development environment:
Apache NetBeans 30 Download
It seems like I've been awaiting NetBeans 30 for ages, but in fact it's only been 9 days since the release was tagged at GitHub. Downloading now!
his father's LinuxSaber
With apologies to George and Marcia Lucas.
[Luke is training with his father’s LinuxSaber ... and a temp file.]
Ben: Remember, Luke: a sysadmin can *feel* the Kernel *flowing* through him.
Luke: You mean, it prompts you for input?
Ben: Partly. But it also obeys your commands.
[Luke attempts to delete the file and a gets a “permission denied” error.]
Han: [amused] Hokey system calls and ancient command shells are no match for Windows Explorer on your desktop, kid.
Luke: You don't believe in the Kernel, do you?
Han: Kid, I've surfed from one end of the Web to the other. I've seen lots of strange stuff, but I've never seen anything to make me believe there's one all-powerful Kernel controlling everything. There are no drivers for *my* HDD. It's all a bunch of registry hacks and nonsense.
Ben: [opens a terminal window] I suggest you try it again, Luke. This time, let go of your mouse and use the keyboard. [He maximizes the window.]
Luke: But with this window maximized, I can't even see the desktop! How am I supposed to delete the file?
Ben: Icons can deceive you. Don't trust them.
Mental model
Mental models are an aspect of human-computer interaction that doesn't get enough attention. Whether you're using existing software or writing something new, having a useful mental model of what the software does (or should do) makes all the difference. In other words, the software shouldn't be a "black box".
The "git" tool is a good example. New users can begin by memorizing the arguments for adding files, committing changes, and so on. But in order to use git effectively, one must distinguish between the database, the staging area, and the working files. One must conceptualize the database as a directed graph of snapshots connected by changes, with one snapshot labelled as the head. Without this mental model, git seems ridiculously cumbersome and difficult. Without this model, small mistakes seem like major disasters.
Similarly, if you begin coding a software project without a clear mental model of how it will work, you'll probably need at least one major rewrite to achieve an acceptable result.
please please please please reblog if you’re a writer and have at some point felt like your writing is getting worse. I need to know if I’m the only one who’s struggling with these thoughts
this is a common phenomenon. the better you get, the more you recognize flaws. the good thing is you can strive to get better. but the bad thing is that you see nothing but flaws. you are actually getting better, but your editing/critical brain is getting tuned up and can see more things to improve. someone post the graph, I can't find it
nvm I found it myself:
"art" can mean any creative endeavor and it definitely applies to writing.
I've experienced these cycles both in creative writing and software development. I think they're less intense for developers because they possess objective metrics and tools for evaluating their product, starting with "does it run"?
We investigated a bug where CUBIC's congestion window became pinned at its minimum floor, causing a performance to plummet. The fix involved
Current reading material
I'm reading this book, written by 3 Stanford professors, about how large tech companies are imposing their values on the United States and transforming its culture. Published in 2021, but now more relevant than ever before.
I P address? baby it’s called the bathroom 😂
writers and artists everywhere all the time
Software developers, too!
*staring at demo code written by the FPGA manufacturer from 20 years ago*
dude it's so clean when can I reach that level of Verilog programming :')
Okay. This is one of those, "How the hell was that ever working?", type situations.
The issue had to do with me not properly validating the return value of open(). I have absolutely no clue how the unaligned msync() had any effect on that, or how this issue didn't show up far sooner.
It was completely unrelated issue, but its inconsistent nature and shockingly far reaching consequences completely threw me off. I'm still baffled how I never noticed that it was even a problem.
Whatever. That headache is over, and I can get on with my life.
The "smoking gun" here was the fact that the start index, which tells the database where the first lode entry is located, was set before closing the file, but cleared after reopening the database. Yet a hexdump showed the entries were still clearly intact.
Once I saw that, continuing with the assumption that msync() couldn't be the issue, that only left the dbfile_open() function. The realization that what I was seeing matched with the file header getting cleared led me to the check for whether the file already existed. It was there that I noticed that, while I was checking for the ENOENT error, I failed to check whether the initial attempt to open the file had actually succeeded or not. I still don't know how I missed that, but considering I'm mostly working on this while exhausted and burnt out from work, I'm also not terribly surprised.
I'm still not entirely sure why the unaligned msync() prevented the bug from showing up sooner. My best guess is, even though it ultimately succeeded, it still set errno, similar to how mmap() does when called with MAP_FIXED, and that cleared the ENOENT code wherever it was coming from, that must have been lingering between closing the file, and trying to reopen it.
It's weird problems like this that are affected by completely unrelated functionality that can run you in circles, and make you want to give up. After getting some rest, and stopping to think about what I knew about the issue, I was able to chase it down until I found the root of the problem. You have to think about what you're seeing, what could lead to that behaviour, and relentlessly follow whatever leads you have to get answers. Debugging can be a slog, but you only fail if you give up. I'm reminded of that every time I find myself in this kind of situation, only to solve the problem through persistence and reason.
popular
My Libbulletjme open-source physics library just reached 100 stars on GitHub.
sipping
Some of the difficulties I'm having with macOS seem to be related to a feature called System Integrity Protect (SIP). SIP interferes with building and installing applications as Administrator.
It's a sensible security feature, so rather than disable it, I've created a non-root account for software development.
During the past week I finally got visualization working for hair simulations. Here's a screenshot I captured today, from a real-time simulation of 9,999 strands, each containing 25 points, at 360 iterations per second. Most of the computation was offloaded to my laptop's GPU.
Now as I study it closely, I'm dissatisfied with the flat-shaded meshes use to visualize the colliders. Always something to improve!
Software used: Linux Mint v22.3, jdk-17.0.15+6, Vulkan SDK v1.4.341.1, LWJGL v3.4.1, Jolt JNI v4.1.0, Sport-Jolt v1.3.0, and Jolt Physics v5.5.0+ (custom-built using clang)
Cem Yuksel created the hair model: https://www.cemyuksel.com/research/hairmodels/
Murat Afshar provided the collider meshes. GIMP v2.10.36 was used to crop the captured image.
A shout out to Jorrit Rouwe @jrouwe.bsky.social for publishing Jolt Physics (on which Jolt JNI is based) and the "HairTest.cpp" sample code from which my (Java) app is derived.
No AI was involved.
And here's the visualization with Phong smooth shading. I'm happy now.
During the past week I finally got visualization working for hair simulations. Here's a screenshot I captured today, from a real-time simulation of 9,999 strands, each containing 25 points, at 360 iterations per second. Most of the computation was offloaded to my laptop's GPU.
Now as I study it closely, I'm dissatisfied with the flat-shaded meshes use to visualize the colliders. Always something to improve!
Software used: Linux Mint v22.3, jdk-17.0.15+6, Vulkan SDK v1.4.341.1, LWJGL v3.4.1, Jolt JNI v4.1.0, Sport-Jolt v1.3.0, and Jolt Physics v5.5.0+ (custom-built using clang)
Cem Yuksel created the hair model: https://www.cemyuksel.com/research/hairmodels/
Murat Afshar provided the collider meshes. GIMP v2.10.36 was used to crop the captured image.
A shout out to Jorrit Rouwe @jrouwe.bsky.social for publishing Jolt Physics (on which Jolt JNI is based) and the "HairTest.cpp" sample code from which my (Java) app is derived.
No AI was involved.
Dario Amodei’s warnings should not be dismissed
This is more than a little concerning. If a tool like Anthropic's Mythos can efficiently find zero-day software vulnerabilities, then software developers will feel compelled to use it (or similar tools) just to keep up.
NPR handled this story competently, obtaining quotes from people at Proofpoint, Corridor, and the Linux Foundation, as well as the lead maintainer of Curl:
Anthropic announced this week that its new model found security flaws in "every major operating system and web browser." Even before the new
The story is growing more nuanced. Organizations are gradually realizing they don't have enough skilled people to keep up with the rising flood of exploits.
Cyberdefenses could be exposed faster than fixes could be deployed.
Okay. Things are broken. Everything is broken. Nothing is working.
This is fine, I expected this. This is why we're running tests. If it worked first try, that would have been astonishing.
It's fine. Everything is fine. I'm fine. We're all fine. It's okay. It's fine. I've dealt with this, I know what I'm doing. It's fine.
