Kudos on your dedication to be self taught, but the questions you raised are one of the things that a university is great for. To make a meaningful contribution in mathematically-oriented fields (such as computational neuroscience), you need to have the following:
1) Access to latest journals and papers: This should help answer question (1), (2), and (3) - use the tools others are using. If you find an open-source tool, that is great. But often, people in the field will expect you to use a standard framework that has been vetted by lots of other researchers.
2) Access to latest data and tools: Matlab costs quite a bit (esp. with all the toolboxes that you might require). Most universities give you the license for free.
3) Like minded individuals are (for better or worse) almost all at universities and research labs and the main interactions come from conferences. Journals are good for non-interactive peer review, but if you want collaborators, you need to head to conferences. This is also where the university name (and financial backing) can help - "Oh, you work with $BigName? I'd love to collaborate with you!"

You don't have to spend a lot of money either. You can take non-degree enrollment (so you can work at your own pace) while still having a lot of access to the tools, data, and collaborators. In addition, you haven't mentioned your background. So you might find it harder or make trivial mistakes that betray your inexperience or out-of-field characteristics. Most graduate (including Ph.D.) students take a lot of classes on basics (at the start) so that they know the vocabulary and concepts necessary to read and understand the cutting edge research. Without that, you are likely too dependent on the tool. I have known lots of people in industry who swear by Matlab (for example), while not realizing how poor it is compared to more sophisticated optimization tools, especially when you get into large data-sets (which I assume you will be involved with).

I read the article, and all I could see is that when you join a group, you get the decryption key for that group - but from whom? If it is automatically done (i.e. Syme holds the key), then it is no more secure to snooping from agencies than any other service (well, except for the fact that it is based in Canada - ah, who am I kidding). What you would need is the group/thread creator send the decryption key directly to the collaborators - which basically means they already need a secure communication medium (sending it over unsecure email is just stupid). Which would then bring me to ask why not just use that medium?

Nope, it is a USB camera - Intel CS110. A quaint old camera; I was shocked that it worked with Windows 8 actually. My primary camera got busted, and I dug this out of my garage. Oddly enough, it works with Skype in Desktop mode, but the turd that is Metro Skype refuses to believe I have a camera.

Seriously - why on earth are you suggesting half those products? While I respect and am aligned with the goals of the FSF, half those "products" are going to change nothing (at best), or backfire (at worst). If you gave a list of decent products, that would be different - I can feel good about giving gifts that people (who don't care about FSF) can appreciate. This just seems like a set of gifts that make me feel good, while my non-techie friends spend hours trying to return the gift for refunds. And what is the point about "free, secure" software, when my family is just going to log on to Facebook and give up their details voluntarily?

Seriously, if this is the best FSF can do, it illustrates the problem: Non-techies really don't care about privacy. Instead, maybe an educational book could be offered? I'd pay for that. These gifts are going to either increase my tech support work or be ignored - the problem is, I want my family and friends to want to use these products; without education, that won't happen. A good reference for the dangers and fixes of proprietary stuff would be useful.

Breakdown of what the page offers:
Trisquel: Modifiable is a selling point? No one in my (or most) family are going to modify the OS in any (significant) way - changing desktop background doesn't count. And I have a intel web-cam from early 2000s that is supported in Windows, but I couldn't find out how to get it to work on my Linux box.
3D printer: Can't comment, so I'm not sure what the free vs. proprietary debate on the printer is about - is it the 3D print file format? Or is it just because the company is evil (hint: my friends don't care, so long as it works. They shop in Walmart and Amazon)
Gift card: This doesn't seem to be a gift card - 20% discount on other merchandise using the membership card? Why not give an apples-to-apples comparison and offer a gift card instead?
Laptop: Well, this is a fair enough. If my friends could use Linux, I might just save the Windows/Mac tax and give them a PC with Linux.
E-book: Again, a fair enough point. But I can give them an Amazon card that will allow my family to get books directly from Amazon and read it on their Kindles. I will not gift them a Kindle, but if they have one, they already don't care. And how do I get Gutenberg books onto a Kindle without having access to their device?
Phone: This is a shot at Apple. Seems like the Android will still tie them to Google's store.
Online storage: Hmm... Can't find any phone app or client application (hint, most non-techies aren't going to use SSH/SFTP - they want something like dropbox or SkyDrive mount).
Media hosting: Most people I know use XBMC, which is open source.

Ahh.. I see you've met my ex

Actually, you are quite a good consumer by most reasonable standards. Not to start a flame-war, but the "must-have-latest-iThing" mentality is in the minority of the people I've met. You have made a lot of purchases, many of them with long lifetimes, so it isn't surprising that you don't get more now.

On the other hand, do you rent/buy movies? Or music? Or games? The first three things you mentioned are largely dead-end devices in themselves. A stereo without music is like a really expensive paperweight.

The only real consumption that I feel forced to take part in is the phone - my last phone was bought in early 2012 (no contract) - but it is getting to a point where planned obsolescence and poor battery life makes me feel compelled to buy one.

Apart from the inflammatory article, I believe there is a valid question to be asked here: how does one identify and catch/correct errors?

In experimental fields, if a result is interesting enough, there will be people who will verify it by trying to repeat or improve the results. However, in more theoretical fields (where computer simulations are the norm), I wonder how well vetted the results are. Especially since many people don't release the source code, and even if they do, it is too large to actually go through and verify each line.

I know of some models (in the aerospace industry) that have been widely used that are flawed in many ways - from sign changes, to impossible geometric configurations. I'm sure that in many other simulation-centric fields, the same problems exist. Often, the results and simulations are just part of some larger methodological contributions - the methodology is still solid, but the presented results are flawed.

Fair enough, and the question is then how confident is your organization in diversification (or more importantly in projecting future trends), and the cost of being wrong. There are countless stories of companies that spent a lot of money developing for multiple platforms, and then being burned when it became clear that diversification was pointless. If, for example, Google released Chrome a short time after firefox started becoming popular, would your strategy have worked.

On the whole, I wonder whether companies in the coming decade will bother to target multiple OS, as hardware/multicore improvements allow efficient virtualization. Either start with an OS-agnostic framework (if it can support your needs), or target one system. Hell, target Linux, because you can give free linux VMs to run under Windows (instead of expecting your Linux clients to pay for a Windows license). Similar to a JVM, but much better because you aren't stuck behind Oracle.

It's a chicken and egg problem - developers won't port unless the demand is there, and the demand won't rise till a lot of people move to Linux, which they won't because they don't have their software for Linux.

I think the whole debate is meaningless - maybe developers will start releasing their work in the form of virtual machines, or users might run virtual machines for their software. A linux base, with a Windows virtual machine (licensing issues with Macs OS on non-mac hardware) for Adobe Photoshop. PCs are good enough that they can take the performance hit of virtualization (except games in some cases). There might be also a few domain specific problems that can be addressed (like color settings for graphics work).

But on the whole, why would developers want to develop for more than one platform, if a single platform can (without too much effort) be used by all possible users?

I was wondering - why did this clam live to a ripe old age? Can all clams live for centuries if they aren't killed by chance (predators, starvation, what-have-you)? Or are clams like most animals in that they grow old and eventually die, even in the best of environments, but it is a really slow process (i.e. like how 20 years to a dog might be 70 years to a human?)

Also, what is it about underwater life that many of the underwater creatures live so long: Maximum life span in animals?

I had a chance to visit ISRO and DRDO (Defence Research and Development Organisation) as a high school student (won some competition), and the difference in the approach was night and day. There are smart people in each organization, but the ISRO people seem to take pride in their work, and are very ambitious to succeed from the top down. They have a very efficient chain of command, and not much politics. They all work towards a common goal - there isn't the head of one lab pitching a mission to Pluto, while another one tries to demand a mission to Mercury.

On the other hand, the DRDO organization is messed up at the top - there are several people desperate to become directors of a lab, and the infighting often results in splintering. There is a 3 mile road where you can drive by six or seven different labs operating independently. While they do some good work, there are a lot of different projects that each lab tries to pitch.

Really? You had to get into aliens/human voyages scenarios for f*ck ups? I'm sure the US' refusal to adopt the SI system (along with illustrious company in the form of Burma and Libera) would burn everyone to a crisp before they got out of the atmosphere. (NASA - Needs Another Seven Astronauts).

No one was paying them extra to deliver on Saturdays. Now, Amazon is footing (a part?) of the bill, and USPS can make money off it. Cognitive dissonance or comprehension-fail?

Also, the proposal to end Saturday delivery failed (first line of the article). And they already (apparently) deliver a some packages on Sundays and holidays for a fee. This just helps them make get a bigger piece of the e-commerce pie.

Actually, India has a very capable transportation network. Their railways are pretty good. Sure, you have the cheap seats that are not clean, but I've traveled a lot on their higher-class compartments, and they were clean and comfortable (air-conditioned sleepers).

The US has one of the worst PUBLIC transportation systems for a western country. Even air transport is horrible for an economy passenger. Frankly, the railway and bus network of India puts the US' public transport network to shame.

There are two ways to improve security - lock out the user, or educate them.

Locking out the user is great - but it only works on NEW products, and if you don't have competitors. The reason it works well on NEW products is that the user isn't conditioned on what to expect. Remember, trying to change how people use their computer is an uphill battle. It works well when the do not believe they have alternatives.

Educating the user is harder, but that is the real fix. You aren't improving security by saying 'As responsible devs, our software won't do what you want'. Instead, make a two minute video showing them how $technology is flawed, and make them watch it ONCE. Then, let the choose whether to block $technology or live with it. Because right now they get fed up with Firefox (NOT Java), and click the little blue e.

And yes, it isn't a great hassle to keep using FF when you allow users to "click to allow $applet". But the pain is that I need to look at the little red icon in the address bar to permanently enable something. You might say that if I can't handle this additional step, I shouldn't be making a choice on whether to run an applet or not (but that is a bad road to head down). You could have just made a popup when I run an applet that says "Do you want to remember this setting?" - it doesn't fix the security problem, but the current solution doesn't either. At least this way, I don't feel frustrated at my browser for someone else's (Oracle, in this case) screw ups.