HN user

pacaro

4,195 karma

github: https://github.com/PollRobots/

blog: http://pollrobots.com/blog

Posts77
Comments1,265
View on HN
pollrobots.com 28d ago

Show HN: Browser based tool for programming ch57x macro-pads

pacaro
3pts0
pollrobots.com 7mo ago

Sierpiński Triangles

pacaro
1pts0
pollrobots.com 7mo ago

Finding Flowers in Chaos

pacaro
2pts0
pollrobots.com 2y ago

Show HN: A tool for rendering and interacting with SDFs

pacaro
2pts0
pollrobots.com 2y ago

Adding a Vim Mode to the Monaco Editor

pacaro
2pts0
pollrobots.com 2y ago

Voxel Ray Casting in Zig

pacaro
3pts0
udel.edu 2y ago

Mathematical Hat in Early Computing Demos

pacaro
1pts0
pollrobots.com 2y ago

Generating a Star Sphere Using Zig

pacaro
13pts0
pollrobots.com 2y ago

Writing a Jekyll Converter for Literate Programming

pacaro
3pts0
en.wikipedia.org 3y ago

William Archibald Spooner

pacaro
2pts0
www.tomshardware.com 4y ago

RISC-V Powered Mango Pi Takes on Raspberry Pi Zero at Its Own Game

pacaro
7pts1
github.com 4y ago

Show HN: A an R7RS Scheme Implementation in WebAssembly

pacaro
84pts11
uk.motor1.com 4y ago

Slo-mo footage of running engine with acrylic cylinder

pacaro
4pts1
differentminds.scot 5y ago

Supporting and Celebrating Different Minds

pacaro
3pts1
www.reddit.com 5y ago

Microsoft Solitaire was written by a summer intern, on their own time

pacaro
15pts4
www.geekwire.com 6y ago

Amazon Employee in Seattle Tests Positive for Coronavirus

pacaro
47pts4
phys.org 6y ago

The Paradox of Unanimity (2016)

pacaro
2pts0
sherylcanter.com 7y ago

Chemistry of Cast Iron Seasoning: A Science-Based How-To

pacaro
2pts2
www.bbc.co.uk 8y ago

The boys on the ice

pacaro
1pts0
github.com 8y ago

Microsoft publishes open source TPM 2.0 implementation

pacaro
2pts0
pollrobots.github.io 8y ago

Strange attractor, 3 dimensional quadratic

pacaro
2pts3
www.thelancet.com 8y ago

Prevalence trends tell us what did not precipitate the US obesity epidemic

pacaro
1pts0
www.bbc.com 8y ago

Robot carpenter makes custom furniture

pacaro
2pts1
www.ncbi.nlm.nih.gov 8y ago

The Big Vitamin D Mistake

pacaro
756pts440
blogs.dropbox.com 9y ago

Dropbox transparency report

pacaro
2pts0
mobile.nytimes.com 9y ago

Travis Kalanick Resigns as Uber CEO

pacaro
2pts0
www.bbc.com 9y ago

The rave that changed British law

pacaro
1pts0
www.bloomberg.com 9y ago

Dropbox Reaches Key Profit Milestone

pacaro
226pts145
www.engadget.com 9y ago

Stanford students recreated a 5,000-year-old Chinese beer

pacaro
1pts1
www.bbc.com 9y ago

Could the world's most expensive spice help farmers in Vermont?

pacaro
1pts0

Can't help thinking of the classic series of userfriendly comics about Vigor, which at one point was implemented as a plugin for nvi

The EU's "E-Evidence framework" allows authorities in any member state to compel entities doing business in any part of the EU to produce and/or preserve communications data, completely by-passing cross-border barriers.

_e.g._ Victor Orban could have wiretapped any communication within the EU. Supporter by an EU directive

I'm getting a mixed reading, the most egregious flags aren't present, but there are a couple.

As you identified there is a lot of parallel sentence structure.

They also have a bunch of lists of 3-5 items which is a classic.

It doesn't have the anodyne rambling never getting to the point style common to LLMs

Dithering as a mechanism to reduce/remove banding can be very impressive.

The color Next machines only had 12 bit displays, 4 bits per channel, but with careful use of dithering it was often indistinguishable from a 24-bit display (so called "true color")

There are so many ways to produce sierpinski gaskets.

It you specify n points and the pick a new point at random, then iteratively randomly select (uniformly) one of the original n points and move the next point to the mid point of the current point and the selected point. Coloring those points generates a sierpinski triangle or tetrahedron or whatever the n-1 dimensional triangle is called

FWIW, the Next color machines had 12bit (4096) color and in practice it didn't make much of a difference. I was using one fairly extensively in the late 90s (by which time 24bit was the norm -- albeit not yet ubiquitous) and for anything other that photo/video work it was perfectly reasonable. It's amazing how far a good dithering algorithm can get you.

The refresh rate is, of course, a different issue

There's even a book "Adventure in Prolog" by Dennis Merritt (ISBN 1520918917)

It's a lot of fun to work through, other prolog resources can be a little dry

For the title holder, in this case the Earl of Lovelace, they are often referred to ("styled") simply by the place name. So after William King-Noel was created Earl.of Lovelace he was styled "Lovelace". She would have been styled "Lady Lovelace" in society, and "Countess of Lovelace" in formal contexts

Just that thinking about the "floppy drive in 1998" question.

I can think of a couple of things at play

The system was likely designed a long time before 1998, because of how this type of contract works

Even in 1998 USB was very new, so thumb drives weren't an option

CDs were ubiquitous, but cd writers were not

Proprietary tech was available, Zip and Jazz drives come to mind, or maybe even minidisc in that timeframe, but any would have been a poor choice in hindsight

Tape would presumably have been considered, but was already notorious for long term compatibility (in 99 I had to restore from a tape backup that was only a couple of years old and we couldn't easily find compatible hardware)

There were other proprietary optical storage media in the market, but they were expensive and aimed at the long term storage market (heaven help you if that's how your archives are stored)

If they needed a cheap, easily written, and distributable media, they didn't have a lot of good choices.

We still don't have a good choice for this if you want your system air gapped. I sure as hell wouldn't let anyone plug a USB device into a safety critical system

I've always thought of this as being equivalent to "work hardening"

My concern with it is more about legitimately old code (android is 20ish years old, so reasonably falls into this category) which was written using standards and tools of the time (necessarily)

It requires a constant engineering effort to keep such code up to date. And the older code is, typically, less well understood.

In addition older code (particularly in systems programming) is often associated with older requirements, some of which may have become niche over time.

That long tail of old, less frequently exercised, code feels like it may well have a sting in its tail.

The halflife/work-hardening model depends on the code being stressed to find bugs

The tech trivia related to this that springs to mind is that the DEC Alpha processor was known internally by the code name EV, which stood for "Electric Vlasic" — although the suits backfit this in to "Extended VAX".

(Vlasic is an American pickle brand.)

I guess that I was taught basic mental arithmetic differently. Multiply by 8 and divide by 5, or (it's inverse) is two single instruction cycle opcodes (or whatever my brains equivalent is).

Regularly needing to translate between sane units and US (and occasional British) idiosyncrasies keeps these mental muscles worked enough that it's mostly subconscious now.

I didn't enjoy rote repetition of times tables and drills as a kid, but it's frustrating seeing my daughter being taught to understand multiplication, and learning "strategies", but struggling with mental arithmetic (I mean she tests above grade level, so I'm not worried, it's just a _get off my lawn_ reflex)

Yeah, this is basically my take too. I had a hardcopy of this sometime between 99 and 02, and read it several times

At the time I was an embedded developer at Microsoft and had been a Windows programmer in the mid 90s. It was pretty clear that there was some dunning Krueger going on here. Neal knew enough about tech to be dangerous, but not really enough to be talking with authority

And in addition, and TFA alludes to this, each time a person receives a transfusion there is a chance that their immune system will generate antibodies to it. That might not impact the transfusion at the time, but can make the recipient much harder to match for in the future.

My partner has had severe anemia, and because of a prior transfusion went from being an easy match (A+) to a being a 1 in a thousand match. The last time she needed blood it took 12 hours to locate and transport suitable blood.

If you read the literature on survival rates for people with very low hemoglobin level, it's all based on post operative recovery (or otherwise) of Jehovah Witnesses

Thanks for the summary. I kinda low-key love the idea of converting floats into a fixed point representation that covers the entire range represented by the float type. I mean the accumulator is only 32 KB, which is likely to be in L1 the entire time on modern hardware, and any given float is only going to need two 64 bit words, + 13 bits (12 bits for offset, and 1 for sign) to be represented in this scheme.

And yet... I worked for a small software company (STNC Ltd.) that was acquired by Microsoft in '99. One of our customers was Psion, and then Symbian, we definitely shipped releases to them after the acquisition, so there was (or could be) Microsoft software on the Psion

On a different not, my understanding was that Psion had paid Cygnus to work on Arm support for gcc. So essentially Psion was responsible for gcc having arm support.

The tool chain was bizarre, writing c++ in the Microsoft Visual C++ ide, and building with gcc. Which was crazy in the 90s

As another comment mentioned, the differ is being done naively w.r.t. the color space. Handling rgb (or gray) values as linear values is usually wrong

Or cut to the chase and use sortition, which will provide statistically correct representation with out even needing a vote

The author mentioned zig. And zig would get this right you can just write `setThreatLevel(.midnight)`

But where zig breaks down is on any more complicated inference. It's common to end up needing code like `@as(f32, 0)` because zig just can't work it out.

In awkward cases you can have chains of several @ statements just to keep the compiler in the loop of what type to use in a statement

I like zig, but it has its own costs too

FWIW iron gall ink is still made and used in the present day. To be legally married in the UK you have to sign the marriage registry with iron gall ink. It's sold as "Registrar's Ink"