I'm entering radiology residency, and I'm very pro-automation / machine learning. There's a contentious debate in the field about whether radiologists will be replaced: https://forums.studentdoctor.net/threads/will-ai-replace-rad... HackerNews is very developer-focused. If you guys saw what a radiologist does on a 9-5 basis you'd be amazed it hasn't already been automated. Sitting behind a computer, looking at images…
As someone who just struggled through writing a DICOM parser, _standardized_ doesn't always mean the same. For more examples, see RETS in the real estate world :)
For someone wanting to begin involving program verification in a practical way to their day-to-day work, do you have any suggestions, resources, or anything?
That depends on your background and your goals. How much do you know about modern testing, abstract interpretation, SAT/SMT solving? In any case, as of Feb 2017, a lot of this technology is not yet economical for non-safety critical mainstream programming. Peter O'Hearn's talk at the Turing Institute https://www.youtube.com/watch?v=lcVx3g3SmmY might be of interest.
>> not yet economical
Don't know much about it, but verum.com claims 50% reduction in development costs.
I'm entering radiology residency, and I'm very pro-automation / machine learning. There's a contentious debate in the field about whether radiologists will be replaced: https://forums.studentdoctor.net/threads/will-ai-replace-rad... HackerNews is very developer-focused. If you guys saw what a radiologist does on a 9-5 basis you'd be amazed it hasn't already been automated. Sitting behind a computer, looking at images…
I'm curious, how hard do you think it is to integrate new tech with the systems currently being used in radiology? Are the file formats and standards largely proprietary/closed?
I'm also doing embedded work, but I don't really see - or expect - performance from the RISC-V cores above similar CPU designs that consume about the same area/gate-equivalents; did I miss some recent results? Of course, freely-available and well-supported CPU IP can be very cool!
Embedded designer here. Came to say RISC-V as well, but not because of performance. It will be because of price. A significant hunk of a Cortex-M die is the ARM licensing fee. If we can drop that? That would be an order of magnitude of savings on my BOM.
>> A significant hunk of a Cortex-M die is the ARM licensing fee.
Can you give a few numbers/guesetimates for common mcu's ?
In the Bitcoin space, I'm most excited about the Lightning Network [0][1] and MimbleWimble [2][3], which are in my view the two most groundbreaking technologies that really push the limits of what blockchains are capable of. [0] https://en.bitcoin.it/wiki/Lightning_Network [1] https://lightning.network/ [2] https://download.wpsoftware.net/bitcoin/wizardry/mimblewimbl... [3] https://bitcoinmagazine.com/articles/mimble…
To add to that, the escalating hashrate war within Bitcoin between Unlimited and Core is popcorn worthy.
It's all a matter of perspective. One could also claim that those who don't acknowledge God are deluding themselves with a lack of theology. I'm not trying to start a debate, but dismissing the beliefs of billions as simply delusion/deception is painting with broad strokes.
Which god? The argument you make could also be made for all people following all the religions aside from the one whose god actually exists. The religions can't all be right, if in fact even one is right. So sheer quantity of believers doesn't work for making a point.
I have used an argument like this and the answer I got(which left me dumbfounded) was like "Everyone believes in some sort of divine power or entity or whatever and the other religions just got their wrong."
They're basically saying that sheer quantity of believers in anything proves that their God exists!
Field: embedded software. To me RISC-V is the most exciting thing for the next few years. The performance appears to be awesome, and free CPU IP will allow more varieties of specialized low-cost chips for specific use cases. It should also have a positive effect on development environments by encouraging wider use of free toolchains.
I'm a complete outsider, but I've always been curious why people didn't build on OpenSPARC. Is it patent-encumbered, etc? Or, is there some technical reason that makes that architecture less worthwhile than a new one?
New aesthetics in web design. With the brutalist movement something new started. People went back to code editors to create websites by hand skipping third-party, non-web-native user interface design tools prefilled with common knowledge making websites looking uniform. The idea of design silos and brand-specific design thinking is dropped: no more bootstrap, flat design, material design, etc. It's like back to the n…
Isn't it just a fad? How usable and readable is the brutalist design? Or what are you trying to maximise other than being different?
Most annoying is the need to write in ES6/Babel today, and all these js hipsters that really believe this is the future of web development. I totally hate Babel with it's dozens of Webpack patches/plugins to make it work. Oh, and don't forget your (Airbnb style) linter if you want to be politically correct. No one needs Babel to write stellar code IMHO. Unfortunately it is not about the quality of the code you write,…
I agree there's a lot of cargo-culting and bandwagon-ism nowadays. It's the result of there being lots of beginners who just want a recipe they can apply to produce a certain result. I don't blame them for wanting that or for being beginners, or for being frustrated at the obvious madness of it all. What bothers me is the apparent lack of interest these people seem to have in understanding the history and trade-offs…
I'd love a factual and accurate report of that history and those trade-offs, even a book-length one on paper. Do you know of one? I'm not trying to do a dissertation on the subject, and it's not my day job, so weaving my own unreliable history out of Google searches and cranky blog posts is not ideal.
As someone who works in the semiconductor industry, one of the most exciting things happening right now is the development and emergence of persistent/storage class memory (PCM/RRAM/3DXP/NVDIMMs). The implications of a persistent alternative to DRAM is immense and besides fundamentally changing compute/memory/networking/storage architectures it will also change programming models and SW stacks as we know them today. This is a topic I feel doesn't get enough visibility here, especially given that support for such technologies has already started getting baked in to Linux and Windows.