Why do you think that source control should be an audit of what each person did, rather than a view of how the software got to its current state?

Yes - posts written in the middle of the night are not entirely coherent :)

Maybe it is not as simple as pick zero / one / two. The purposes of writing software for research and engineering software for reuse are so different that it doesn't make sense to try and compare them. Going back to the summary:

What many academic researchers fail to understand is that this specialization problem is not unique to research projects. Most software developers will seek to provide an adequate solution to their specific problem, as quickly as possible. They don't seek to build a perfect, all-purpose, tool set that can be reused in every conceivable circumstance.

No. What the author of the article fails to understand is that software is not the point of research - it is a side-effect, and I say that as someone whose field is CS. We do not write software in academia because we want the software - we simply want the data about its behaviour that we can get from it. It doesn't matter if business / hobbyists / academics have in common an approach that builds software for the least effort. In the first two cases the software is being written because there is a need for it to be used. In the latter case it simply needs to exist in some form long enough for some data to be collected and then it is obsolete. This difference is purpose is so vast that it renders the rest of the argument in the article as not even wrong.

The period is directly proportional to the loveliness of a summer's day, perchance?

This system creates the illusion of intelligence. We know from fMRI that "free will" does not exist and that "thoughts" are the brain's mechanism for justifying past actions whilst modifying the logic to reduce errors in future - a variant on back-propagation. Real-time intelligence (thinking before acting) doesn't exist in humans or any other known creature, so you won't build it by mimicking humans.

So how do you account for effortful thought or planning? It is true to say that there is no thinking before reacting, but to claim that there is no thought before action is absurd - how do you explain extended endeavours such as writing a book over the course of a year? That must be one hell of a chain of unlikely events that caused that number of reactions, which were combined without thought to produce a coherent written narrative.

Your other claims that memories do not exist and are synthesised on demand are interesting - do you have any references?

Very apt - the ambiguity is delicious.

The atmosphere (which the smog is within) is part of the earth so by the time the sunlight reaches the smog it has already arrived. The real question would be if the particles in the smog reflect the light or absorb it as heat.

You haven't actually represented it as a switch-case. As the other commenter points out you have tacked a redundant switch-case onto the code:

  sides[i] = dot < -LIGHT_CLIP_EPSILON ? SIDE_BACK : dot > LIGHT_CLIP_EPSILSON ? SIDE_FRONT : SIDE_ON;

CS instructor eh?, wouldn't let you cover my class for me... :)

Thank you. That is exactly what I was looking for, and so with a straight face I must admit that your googling skills are somewhat superior to mine. Cheers for the link.

I have used google. I've looked at the results. Hence the list of criteria that they don't match. Feel free to try and prove me wrong by posting a link to something that does match what I listed...

About the size of an RFID. The temperature sensor was an off the shelf MEMS in an IC package, there was the main PIC as a processor and some other odds and ends to link it to the bus. We weren't using Ziggbee, but that is a single package now so I'm thinking the whole thing could be a lot smaller than a cell phone. The battery was just a watch battery (don't know what they are called, the little circular ones). The power draw from the processor, sensor and bus was practically zero. Essentially it is a question of how big a battery is needed to fire up the wireless once a minute and squirt out a packet.

The link is still broken so here are some semi-random thoughts on what kind of remote control I was looking for a few days ago. I want a reasonably accurate temperature sensor, not +- 1 or even 2 degrees C. I want +-0.1 degrees with a update frequency of about once a minute. Wireless with a 50-100m range to a base-station that can then log the data. I don't want some kind of "cloud service" where the base-station uploads my data to the maker's server. I don't want a closed proprietary interface. I want the battery in the sensor to last at least a year, and the sensors should be dirt cheap (less than $20) so that I can put one in every room in the house and map heat flow through the rooms.

Why does this not exist? We were building these kinds of devices in a lab ten years ago, I assumed they would have been commercialised by now. Instead the home automation products that I found were outrageously expensive and limited (proprietary interface to a private server, all access via web). There are no technical limitations, and the parts are cheap enough to make a decent margin on $20 for a PIC microchip, a Zigbee controller and an IC for temperature. It does seem weird that the market that exists for home automation is so distorted.

Are you assuming that Schrodinger never watched his collection?

Making the install size of the system smaller enables new applications and ways of using the system. At the moment an operating system is a giant monolithic install tied to a signle machine. I would prefer a single install shared across all of my devices. To make this work the OS would need to be synchronised / reconciled across a network connection on demand. Like a cloud OS but not running on public hardware - just migrating between devices that I own. As size as a direct impact on performance (time to start up, to reconcile) the install size could never be small enough.

There are smaller devices than you acknowledge. Memory and storage are the dominant costs in those devices and reducing the need for both enables cheaper classes of device. The Pi is an interesting machine for $25, but what happens when a desktop-capable machine costs $5, or $1?

Lastly, flash is nice but it still is not as fast as DRAM. Having an entire OS install cached in memory has advantages for performance.

When you have a filesystem that understands hard links, deduplication is still required to find files that have the same content and link them together. You are possibly thinking of a filesystem that hashes contents to decide on storage locations.