rationalthug

On my personal list of the top ten science fiction novels of the last twenty years. Still re-read it every now and then.

Can you be more specific about how how her viewpoints create problems for you as a science educator? (I am speaking specifically about her content relevant to physics/particle physics/cosmology.)

Generally, she is careful to point out when she is voicing views based on her own preferences, speculating about things that have no observational data to back them up or giving personal hot-takes on various topics.

For everything else, aka, the things she accepts provisionally as true, she appears to favor the least speculative among current solutions/theories/approaches in physics. Much of her popular work is focused on critiques of the many speculative musings of physicists who, in her opinion, have moved into a kind of post-science research program.

Her acceptance/popularizing of the standard model and of general relativity as humanity's current best models of the universe indicate that her viewpoints tend to be highly conservative and evidence-based.

Again, most (all?) of her musings about things that might be true or things that might be interesting directions for research are clearly called out as such.

Aerospace specific software that goes to space or code that just happens to go to space as part of bigger projects?

Here are a couple of the former:

RTEMS https://www.rtems.org/ Core Flight System (cFS) https://github.com/nasa/cFS

NASA has a lot of open source projects including a bunch that don't "go to space" but are used in space related projects (check each project for contributor guidelines):

https://github.com/nasa/openmct - web based mission control software https://github.com/nasa/trick - simulation for for all phases of space vehicle development

You should check out their github organization for all available projects (https://github.com/nasa).

Also, European things as well such as the ESA's github organization: https://github.com/esa, and Germany's DLR: https://github.com/DLR-SC.

Another good group to check out is TU Delft: https://github.com/tudelft, https://github.com/tudelft-iv and others.

Not to mention a bunch of open source CFD projects on github that are widely used and other similar tools. Again, not strictly software that goes to space, but you will find more of that too if you search for projects related to those mentioned above.

For those interested in CAN bus for UAVs, there is https://github.com/uavcan as well.

For node, there is https://github.com/fluent-ffmpeg/node-fluent-ffmpeg.

I’m curious to know if they ever looked at https://github.com/Microsoft/napajs from Microsoft or any other in-nodejs solutions to deal with the need for multithreading vs launching multiple processes...

For anyone interested, here is a sampling of available ML libs in JavaScript. The first two, including tensorflow.js, are sub-libraries of the major ML projects, and the others are custom JS libs:

[tensorflow.js](https://github.com/tensorflow/tfjs)

[mxnet.js](https://github.com/dmlc/mxnet.js/)

[propel.js](https://github.com/propelml/propel) - this uses tensorflow.js under the hood

[ml5](https://github.com/ml5js/ml5-library) - also uses tensorflow.js

[webdnn](https://github.com/mil-tokyo/webdnn)

[brain.js](https://github.com/BrainJS/brain.js)

[mljs](https://github.com/mljs/ml)

[synaptic](https://github.com/cazala/synaptic)

edit: added the ml5 lib mentioned in a subsequent comment.

How does Cloud Functions for Firebase handle node modules with native code add-ons?

You do realize that Python was not only built to appeal to "non-devs" as well ad devs but, does, in fact, enjoy some of its strongest support from them- scientists, engineers, statisticians, etc., who do not consider themselves to be good developers. And, man, they can develop some crappy code/systems in Python too.

I like Python too (as well as node,) but your characterization of node is over-broad and, imo, inaccurate.

Depending on your use case, you can also use ES6 features on AWS Lambda by embedding another node version in your zip package. You then use the lambda supported version of node to launch/bootstrap your chosen version when the lambda function is activated. Of course, Webpack still supports more ES6 features than node at the moment, but the gap is rapidly shrinking.

On a somewhat interesting note: transferable objects work similarly in web workers. There have been complaints, of course, that this makes it impossible to take advantage of shared memory in cases where it provides clear performance benefits.

I think this is the best article I've read about stream processing yet. Love how all of the related concepts are tied together and how concepts in one form of stream processing naturally fall out of the structure preferred in another form of stream processing.

Great explanations and insights.

The author's terrorism concerns are way overblown and the reasoning supporting them specious, but there is a more basic issue with hypersonic point-to-point transportation systems outside of the additional arguments related to current air travel, at least for vehicles designed to go suborbital/exoatmospheric:

Once we are talking about traveling to a destination and not simply going "straight" up and down, the costs, maintenance issues, safety requirements, development time, thermal management, mission management, guidance systems, etc., get within spitting distance of actually developing an orbital vehicle.

Future technology improvements and cost reductions cannot be ruled out, of course, but given the basic physics and what we know of the near future of aerospace technology, it seems a stretch (at least) to expect a profitable, safe point-to-point suborbital system to be a viable undertaking.

And once we are talking about orbital systems, where the cost and risk/reward equation at least seem to be more desirable and unrelated to point-to-point travel, the question of suborbital point-to-point fades into the background.

In a similar way, a separate marketing tech team can also create tension with the rest of the developers when marketing gains control over various parts of a site/service due to clear success in moving the revenue needle. There's almost nothing more frustrating for a developer than starting on a great product concept, but being told to drop it because marketing doesn't want such a product or feature to interfere with what they're doing (A/B testing, etc.)

Great work. This is exactly what I had been idly wondering about yesterday as I was setting up a new project that uses PM2. Does this obviate the need for pm2-interface, or will that project still be developed with a focus on the RPC angle?

There's also http://appleseedhq.net

How does the binary support/performance and streaming support (with socket.io-stream) compare to other web socket libs like https://github.com/binaryjs/binaryjs, https://github.com/einaros/ws and https://github.com/maxogden/websocket-stream ?

I already commented on a different thread about the Flux architecture, but I'll state it again here: Flux seems like a slightly more constrained, perhaps less effective way of doing the same kinds of things that state machines do.

Is there something about Flux for typical use cases that makes it a better solution?

After reading the above article, I can see how Facebook ended up with their current approach, but it seems to me that Flux is a roundabout and possibly inferior way of using state machines to organize application logic and user interface communication. For instance, statements like "The Dispatcher can be made to reject actions while it has not finished processing a previous action." indicate (to me, at least) that Flux could likely be more efficiently coded and used if it were based explicitly on hierarchical state machines.

This is fantastic. I've been reading through bunches of patents lately doing research for a project and this actually made laugh out loud.

Why do authors and publishers of articles like this, which invariably turn out to be about medical knowledge/studies/research, resort to using the misleading, incredibly broad word "Science" in their titles? All of "Science" has lost its way? Really? Phyics? Chemistry? How about this for a headline: "Journalism has lost its way"? The article could then simply be a list of all the sensationalist, purposely misleading crap that's published in major publications. Long list.

While cutting time in the air by half may not make a fundamental difference, cutting it by five or six times could. Point-to-point sub-orbital services have the potential to do this, though there are challenges for this type of flight that bring its difficulties closer to orbital operations than the (mostly) straight up and down trajectories that will be flown by Virgin Galactic, XCOR, etc.

Nicely done. 3D Product previews seem like a good use of CSS transforms.

I had a negative reaction when I first saw the string literal syntax, but after reading through your docs I can see the appeal. Using the Harmony syntax seems like it could be a big win. RequireJS is the module loader I currently use, but I'll give the source of Shepherd-js a once-over and if it looks promising I'll give it a go.

But it's unclear if this is what "everyone else" wants as well. The most interesting group, I think, is not the mostly tech-illiterate who want to store their garden pictures and videos in the cloud, nor the tech-elite who are willing to trust online storage with large swaths of their lives and businesses.

It's the group in between that is the most interesting to these companies right now. Those who are relatively well informed about major technologies and their uses. Do they really want to have their information mostly stored off-site, or do they simply want larger and larger local storage mirrored by online back-up for peace of mind and easy sharing?

I would guess that this is the largest (apparent)consumer market and one of the reasons why this is where most of these companies have their focus right now.

I'm almost ashamed to admit I'd never heard of either of these before. Will definitely check them out.

Cool. I'd guess that SEO/SEM for a domain name business would be incredibly important, but I could be wrong, and your real-life experience would be interesting.

I use TrueCrypt on my Dropbox volume, but it really would be nice to have an integrated solution. Many non-technical users may not realize they care about this, but they do.

I suppose a next step would be to perform a brain scan before someone starts using Facebook, then several over the course of their subsequent use. The researchers would obviously need to do their best to control for confounding factors, but it seems like a reasonably interesting first study could be done this way.

Great list of bootstrapping advice. You haven't included a section on pricing, though, and I'm curious what quirks, problems and insights you've encountered. It seems as though the domain name business might have some unique challenges for a start-up...