sklogic

67 ms·

691.
by sklogic 11y ago
Nice! But I suspect ot would have been much easier with the OpenCL driver and the current LLVM GCN backend.
692.
by sklogic 11y ago
I played a bit with the "more faithful" genetic algorithms, got some promising results (getting out of the local minima better). The key is to use hierarchical encoding in your chromosome: instead of directly encoding traits/
693.
by sklogic 11y ago
Another unwarranted push of the higher education, which is by definition all about science and ivory towers, into the totally irrelevant "employability" bullshit. Please stop undermining the value and the role of the universities.
694.
by sklogic 11y ago
Even before Alpha, something like a VAXStation (think of the 3100 line) would have made an amazing alternative to PC.
695.
by sklogic 11y ago
A killer feature of Mendeley for me was its ability to extract metadata from PDFs. I remember that I could not find a comparable functionality in Zotero last time I evaluated it. Is there any way to do it now? EDIT: found it! Great!
696.
by sklogic 11y ago
A Mill-like architecture would require a huge register file (alternatively, a huge shift register). Not sure it's a good idea for a TTL design. Accumulator architectures are better suited.
697.
by sklogic 11y ago
Shit. I'm using Mendeley and did not know this. I'll stop using it at once. Any suggestions for alternatives?
698.
by sklogic 11y ago
Exactly. Tinkering with things is not any different from any other games children play.
699.
by sklogic 11y ago
The main question is how hard would it be to install a proper Linux on this device. ChromeOS is useless.
700.
by sklogic 11y ago
Good luck getting to that tflop upper limit on anything but the most trivially parallel tasks. Phi got much more flexibility than the GPUs.
701.
by sklogic 11y ago
Cuda computational model does not fit Phi well, you need a different approach (and, ideally, a different language).
702.
by sklogic 11y ago
Why would you care about selecting the branch in compile time? We're talking about language features semantically equivalent to pattern matching, and, turns out, there are none. As for the importance of the exhaustion check, see the b
703.
by sklogic 11y ago
type 'a myOption = Some of 'a | None let f (Some x) = x*x Gives this output: Warning 8: this pattern-matching is not exhaustive. Here is an example of a value that is not matched: None
704.
by sklogic 11y ago
I have not published the language yet, have to produce some nice looking examples first. But you can take a look at my approach to extensibility in general (the HDL language is using exactly the same thing) and some of the earlier mixing Ve
705.
by sklogic 11y ago
> Exactly as with pattern matching With pattern matching you'll get static checking if your matching is exhaustive. Impossible with an if ladder. Therefore they're not semantically equivalent.
706.
by sklogic 11y ago
This is actually something I am currently working on. An extensible HDL which allows to gradually add abstraction layers (and operate solely on any chosen level), while keeping an ability to express anything down to gates.
707.
by sklogic 11y ago
They can call it any way they fancy, something like "Majestic Royal Type Procession" would have been equally ambitious. But in a context of ML, type inference is a Hindley-Milner inference or an equally powerful equation solving s
708.
by sklogic 11y ago
You can also do it in Brainfuck if you like, with an equivalent code bloat. Pattern matching is an order of magnitude more readable and dense than the if ladders. As for the hiring, it is a silly idea in general to seek "programmers in
709.
by sklogic 11y ago
> var a = 0 This is not a type inference. It is a mere type propagation.
710.
by sklogic 11y ago
> that can be easily replaced with similar constructs Mind giving an example? > Using a not very popular programming language can actually make it harder for you to hire people. Do you really want to work with people who can't l
711.
by sklogic 11y ago
Imagine you have to write an optimising compiler for some exotic DSL, targeting a custom embedded platform. All the hipster "frameworks" and "vast number" of irrelevant shitty libraries won't help at all, while patt
712.
by sklogic 11y ago
> Native and 3rd party libraries+frameworks far outweigh usability and productivity gains It heavily depends on what you're doing. You might be surprised, but there is a lot of projects which do not rely on any 3rd party code and
713.
by sklogic 11y ago
In a high level language? Absolutely. In C? No, thanks. There is a huge semantic mismatch with the highly parallel nature of FPGAs. A high level language suitable for HDL generation must expose much better abstractions for parallelism. CUDA
714.
by sklogic 11y ago
Generating HDL out of C (or OpenCL) is a very bad idea anyway.
715.
by sklogic 11y ago
Property developers would be puzzled by this suggestion.
716.
by sklogic 11y ago
To be honest, I'm not quite aware of the current school requirements here, in my days all reports were hand written. But I've seen quite a lot of young people (here in the UK, do not know about the other countries) who obviously g
717.
by sklogic 11y ago
> GUI text editors Emacs is a GUI text editor too, btw. I can be wrong, but my impression was the opposite - all the new tutorials I saw recommended one or another starter pack with all the bells and whistles out of the box. > while
718.
by sklogic 11y ago
> more familiar To whom? There is still a huge number of incompatible text editor cultures. Mac bindings, Microsoft bindings, Wordstar, etc. > better defaults. There are dozens of "better" .emacs pre-cooked packages. Choo
719.
by sklogic 11y ago
Did I miss something and Sublime somehow got some advantage over Emacs? When did it happen and what kind of advantage, exactly?
720.
by sklogic 11y ago
Nothing can improve simplicity and readability more than macros. They make sense most of the time , almost always, not just in some edge cases.

More ›