rdw

16 ms·

121.
by rdw 12y ago
I was curious about Rust's support for within-application custom allocators, it looks like this is still under discussion: https://github.com/rust-lang/rfcs/pull/244
122.
by rdw 12y ago
MedXT is YC 2013. Congratulations to Reshma and Cody, they've been working extremely hard on producing a very impressive product.
123.
Making of: draggable handles(redblobgames.com)
5 points by rdw 12y ago|0 comments
124.
by rdw 12y ago
Very impressive stuff, but looking at the models raises a question in my mind. Why do the parts look like they have about an inch of supporting, sprue-like structure beneath them? Wouldn't it be better to have them be closer to the s
125.
PlotDevice: A Python-based graphics language(plotdevice.io)
1 points by rdw 12y ago|0 comments
126.
by rdw 12y ago
The biggest turnoff to me is the sheer quantity of code (regardless of the fact that it can all be crammed into a single header file). It seems much more difficult to fix a bug in it than in something like UnitTest++. That said, inline set
127.
by rdw 12y ago
This looks hella good, I particularly am intrigued by the nested setup/teardown system. The natural assert syntax is interesting, too, amazing what macros can accomplish. Appropriately, the first question that came to mind was, "
128.
by rdw 12y ago
This is the basis of the Voxel Farm engine, which Miguel talks about in this early blog post: http://procworld.blogspot.com/2010/11/opencl-first-iteration...
129.
by rdw 12y ago
Posing him as an opposition to Snowden is not right, because Tye was only able to blow the whistle himself because of Snowden's leaks. "Tye said he never would have even learned about 12333 and NSA capabilities short of Snowden’s
130.
by rdw 13y ago
Won't take off in the marketplace? Seems plausible. Doesn't produce novel experiences? Already false but maybe this can be spun convincingly. But the technology can't possibly work ? That just seems silly.
131.
by rdw 13y ago
Thanks for the gracious reply! I'm sure that you and everyone working with you at Joyent is a fantastic individual. And yeah, just illustrating how even the best-laid plans of the best of us can't always work out great for all cu
132.
by rdw 13y ago
Really appreciate the open source software, but I recently had some trust issues with Joyent the service. Cloud servers switched from tiered transfer rates (with a generous free tier) to on-demand pricing, with only a month of notice. It
133.
by rdw 13y ago
I think that there wasn't as much demand for batteries in the past as there has been this decade. Think of all the things that need batteries now that didn't exist in quite as large numbers 10 years ago. Laptops, phones, table
134.
by rdw 13y ago
Doesn't this defeat one of the purposes of Javascript CDNs, that the user already has the exact url cached on their machine?
135.
by rdw 13y ago
I realized recently that the reason that there is no way of mocking C is because of a weakness in the language implementation/design. In Python, it's easy to mock and unit test globals, because in the test you can just overwrite
136.
by rdw 13y ago
Whether or not it can buy happiness, money can buy adulation.
137.
by rdw 13y ago
Hah, I guess I bought into their marketing, then!
138.
by rdw 13y ago
That is an excellent explanation! What it suggests to me is that the problem with patent trolls is perhaps not so much with the patent system itself but with the specific mechanisms through which patent lawsuits proceed. Kind of like how s
139.
by rdw 13y ago
Intellectual Ventures actually does create the patents. We think they're for obvious, trivial ideas, but they definitely create them. (I guess they don't actually sue other companies directly, either.) My understanding was that
140.
by rdw 13y ago
What's interesting to me about this story in light of current events is that Johnson Research and Development Co., seems in many ways like an NPE. He doesn't make products directly, he simply licenses the patents to other compani
141.
by rdw 13y ago
Ah, my brain was somehow not seeing that in the article, but I see it now that I'm looking for it. Thanks!
142.
by rdw 13y ago
"you can't have more than 64,000 objects in a folder in S3 - even though S3 doesn't have folders." Is this for real, or are these stories made up? All documentation I've read about S3 suggests that it does not have
143.
by rdw 13y ago
Also of note: there is the Nordic Game Program, which encourages the development of original game projects. http://www.nordicgameprogram.org/?id=23 (for all Nordic countries, not just Sweden)
144.
by rdw 13y ago
How can you not be calling them "whackages"? You're sitting on a gold mine!
145.
by rdw 13y ago
It's either reflected or turned into heat within the panel.
146.
by rdw 13y ago
I really appreciate they have a "reproduce the results" section. We need a lot more of that kind of thing in CS research.
147.
by rdw 13y ago
That's a clever idea, but I think that most users would be disconcerted by the sheer amount of information that's collected by most web sites already. It would raise the question, "why do you need to collect all that?"
148.
by rdw 13y ago
Interesting, that surprises me as well. Thanks for sharing!
149.
by rdw 13y ago
How would you compare this to LiveLoop?( http://www.getliveloop.com/ ) It looks like a very similar approach, is it just sans the real-time aspect? (I know the LiveLoop folks, so this caught my eye.)
150.
by rdw 13y ago
Thanks for the supplementary links, they really clarify things for me. I think the key high-level decision they made is to ignore per-transaction latency in pursuit of higher throughput. A consequence of that decision is that the applicat

More ›