Hacker News new | past | comments | ask | show | jobs | submit | juliangamble's comments login

Quibble: Both Clojure and Scala have a Software Transactional Memory implementation, and the original Clojure Ant demo showed this.


I did the tour of Bletchley Park today and my Tour Guide said he'd met Betty Webb, that he mourned her loss, and that when he had met her at a reunion, she had remained tight-lipped about what her work had been on.


> These “limitations” inhibit truly intelligent behavior in machines, LeCun says. This is down to four key reasons: a lack of understanding of the physical world; a lack of persistent memory; a lack of reasoning; and a lack of complex planning capabilities.


aligns with (or is based on) Demis Hassabis' assessment from yesterday on missing cognitive capabilities for AGI: long-term memory, reasoning, hierarchical planning. He then goes on to suggest scientific creativity may be essential.

https://www.youtube.com/watch?v=yr0GiSgUvPU https://news.ycombinator.com/item?id=42817089


The University that hosted "The Little Schemer" and all its derivatives. Those books were a delight and taught me to think about programming in a way that other languages had not. (Even if I didn't go to that University). https://prl.khoury.northeastern.edu/teaching.html

I had been to a Scheme conference in Washington, adjunct to the Clojure Conference one year, and it was attended by many undergraduates from Northeastern, (and the authors of those books that I got a photo with.)

I have to feel sympathy for those undergraduates I spoke to. They gave a strong feeling, even then (8 years ago), that it was time for the University to move on in language choice.

I had a similar experience in the late 90s when the world was picking up Java, and our University insisted on teaching in Eiffel.


Like a @Nanoraptor concept art, except real…


I’ve tried different part detectors, and nothing has been as good as the Google Image search (download the Google App if you’re on an iPhone.) I’ve found it surprisingly good at identifying LEGO parts.


There is some code to generate Lego images from Lego 3d files here and use it to train models:

https://github.com/jtheiner/LegoBrickClassification

It is based on this post (which others have mentioned)

https://jacquesmattheij.com/sorting-lego-the-software-side/


> Acorn did ship a computer with the 65816—the Acorn Communicator—but when Sophie Wilson visited WDC in 1983 and saw Mensch and crew laying out the 65816 by hand, it struck her: if this motley crew on a card table could design a CPU, so could Acorn. Thus Wilson and Steve Furber forged their own CPU: the Acorn RISC Machine.


The first commercially available Acorn RISC processor was released as a co-processor for the BBC Micro. Acorn always had processors on the mind it seems as the Tube interface and protocol [1] is solely for co-processors.

There's an excellent Rasperry Pi based project, PiTubeDirect, which emulates the ARM and many other co-processors on original Acorn 6502 based hardware; Atom, Electron, Micro and Master [2]. The original expansion hardware is, as expected, incredibly rare and valuable.

[1] https://mdfs.net/Info/Comp/Acorn/AppNotes/004.pdf [2] https://github.com/hoglet67/PiTubeDirect


Watching some Sophie Wilson videos over the years, she is a tremendous person.


I’ve heard some claims that to get closer to brain-like energy efficiency you’d need to use a spiking neural network https://en.wikipedia.org/wiki/Spiking_neural_network


The author spends lots of time modelling axial tilt to represent the seasons.

Then there was this gem:

> I also used this program to create digital building instructions. My takeaway from this is that it works really well for small models, but it has serious usability problems on larger models. When you make changes to the model or change an early step in the instructions, it will often break the layout of the instructions in many places so that most of it needs to be redone. This was a lot of tedious work, but I ended up with a 264 page, 436 step instructions PDF.

Wow.


That actually doesn’t seem that bad for a model this complex. The larger sets can have well into the hundreds of steps with manuals reaching into several hundred pages on 8x10 paper.


I used a Lego product this past week and the instructions were actually omitting 1-2 steps explicitly and expecting you to see the difference between the current and previous figures in the print.


In the old days, "spot the difference" was the standard practice in Lego instructions [0]

0: https://www.toysperiod.com/download.php?file=h4d4k5w544a4h4n...


As it should be. I credit LEGO for my acing the ASVAB spatial reasoning sections.


Consider applying for YC's Summer 2025 batch! Applications are open till May 13

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: