That's not quite true; even without a "stop the world" collector, you can still look at the amount of time the parallel GC thread is busy. In fact there seems to be quite a bit of information available with tools such as jstat. Perhaps this could help you better understand actual heap use (this is also a problem I face at work -- processes that run out of heap after many hours, expectedly... argh).

Some info about jstat: http://www.cubrid.org/blog/dev...
Info about G1 (although you already seem very familiar with it): http://www.infoq.com/articles/...

Compared to Commander Keen? Yes.

So if it's "hugely unlikely" to reduce the UK's output, when their per capita emissions are higher than China's, how can we expect China to do it?

Surely it's both fairer and more realistic to treat the targets on a per capita basis and not penalise China for having a large population.

I love it. Thank you! That's going to be my new tactic from this day hence!

A quick web search will solve that for you. The phrase has been bandied about for quite a few years now, although many people disagree on the topic.

While I certainly agree that C is a significant (and useful) abstraction from the machine -- or more specifically, the assembler -- level, I think you've glossed over quite a few things that are much closer in C to how it works at the machine code level, compared to most other languages. Some that spring to mind are:

1. Pointers and pointer arithmetic -- and not just for fun; you need to use them to get things done.
2. Array indexing -- this is little more than syntax sugar atop pointer arithmetic. Indexing element i of an array is equivalent to referencing *(array_base + sizeof that type * i), and accordingly there is no bounds checking etc.
3. No real strings -- just arrays of characters which suffer from the same problems above.
4. No garbage collection or even refcounting -- if you want to return a string value from a function, you need to either take a buffer (and, ideally, the size of the buffer since you can't automatically determine it) and store the string there, or you need to dynamically allocate memory and return a pointer. Whoever called the function then needs to "free" that pointer later. That's some pretty low-level stuff you have to do just to pass some strings around!

These are just a handful of many things you need to live with in assembly AND in C, unless you use some non-universal, non-standard library for e.g. strings or GC (Hans Boehm's drop in one seems pretty good though). Additionally, C forces you to declare the types of each variable and function, yet cannot properly enforce type-safety and has no type inference which would have made the job easier.

C is undeniably less low-level than assembly, I consider C to be a low-level language in an arbitrary line-in-the-sand sense. Of course, my view on this is no more valid than yours since how we define "low-level" as some absolute marker is pretty subjective!

C is most certainly a low-level programming language. There's a reason people call it "portable assembly language". Of course, as with almost all programming languages, people build useful abstractions in C to bridge the gap somewhat. But that doesn't make C itself a high-level language, any more so than does the use of functions and macros to increase the expressive power of an assembly language.

JPEG is quite robust to corruption, and even PNG's lossless compression seems to be tolerant of a few stray bytes. However, encrypted files would probably be badly damaged by this sort of corruption.

This is the kind of situation where some form of transparent, redundant error-recovery system is extremely important. I'm sure that in the medium term future (after everyone is using SSDs and the cost/capacity ratio falls much further) some kind of RAID setup will be the norm and these kinds of problems will become vanishingly unlikely.

Actually you are completely wrong about that. The data suggests that no one can demonstrate even SHORT term photographic memory.

You stated that you can "look over a page" and "load" all the information into your memory before comprehending it. This is an utterly ludicrous claim. You also stated that you can read at over 1000 WPM with very high comprehension. Also a laughable claim that is not supported by any scientific studies.

Agreed. There's only been one scientifically controlled study that suggested anything remotely like an eidetic memory, and the subject ("Elizabeth") later married the scientist running the study. Then they refused to participate in any further experiments. So, in effect, nobody in the world has ever successfully demonstrated either an eidetic memory, or (much less) an ability to read significantly faster than average while maintaining decent comprehension. Any anecdotal arguments (for example, Murdoch5's frankly laughable claims) otherwise are almost certainly pure Munchausen-esque nonsense and should be disregarded with prejudice.

Nobody in the world has ever demonstrated such a capability under scientifically controlled conditions. All the evidence gathered to date suggests that you are lying. And that's a bit sad.

https://en.wikipedia.org/wiki/...
https://yourlogicalfallacyis.c...

Although it's certainly important to consider multiple viewpoints in a debate, it's wrong to assume that the truth must lie between two extreme positions.

Personally, I think the guy risked everything to expose what is clearly (to most people in the world) a bad and pointless system. He's certainly more deserving of a Nobel peace prize than Obama, who hasn't even shut down Guantanamo Bay yet. The ironing is delicious.

But of course, it's become industry-standard best practice!

Apparently logic puzzles don't provide much benefit in brain function, even though it seems intuitive that they should. Strangely enough computer games of various sorts have been shown to improve cognitive abilities. Playing tetris for a little while every day actually resulted in a measurable increase in thickness of cortex tissue in one study, along with improved language skills of all things. Playing first-person shooters helped make certain parts of the brain process things more efficiently.

Then, for working memory and (I think) executive function, the adaptive dual n-back (DNB) has shown to be quite useful, although the effect size might be smaller than initially reported.

You're absolutely right that proper exercise, diet and sleep are by far the most important things to have. Having poor quality sleep in particular has an enormous negative impact on long-term memory and can lead to atrophy of the hippocampus and parts of the prefrontal cortex. So I'd recommend spending a lot of effort on really fixing your sleep as it's worth proportionally more than probably any other intervention, and it's easy to keep fixed once you've gotten into a good routine. A good routine is at least 7.5 - 8 hours of decent uninterrupted shuteye at consistent times. Dietary and behavioural changes can also help improve sleep and ensure that you get good slow-wave activity (this is the part that usually suffers most as we age, and is linked with memory deficits). Try to measure your activity with a sleep tracker, and see what's going on.

I don't think that's true. Maybe the inverse - 99% of the populace is actually quite intelligent and could do very interesting things if they weren't forced into a career of "make-work" to support themselves.

It appears that you either did not read, or did not understand, the comment to which you replied.