gprisament

I've found linux to be the best environment for doing programming work. The command line is a powerful tool. The key is scripting repetitive tasks, so you only have to ever do them once. Most stuff in windows can't be scripted very easily.

>> Perhaps both ideas are too saturated?

I'd say so. Unless you have something that really differentiates your solutions I'd try to come up with a more unique idea.

There are a number of serious problems with this vision:

1) Of course the digital components of cameras (storage size, sensor resolution, processing power) will continue to improve at Moore's law pace, but optics is quickly becoming the limiting factor. In particular, diffraction limits sharpness as you increase depth-of-field (a result of the small aperture size) and the limited number of photons in dark scenes limits our high-ISO potential and resolution on small sensors.

2) "...the ability to keep everything viewable in focus at the same time".

Even if you could do this you wouldn't want to. Depth-of-field and focus point are some of the most important creative decisions a photographer makes per photo, and since they are "3D" phenomenon the effects cannot be accurately simulated during post-processing. (maybe stereoscopic cameras would be able to but then you need two lenses, which would add cost & weight in comparison to a "single" lens model)

3) "But perhaps the most radical thing about this camera is that it's really a camcorder. Rather than take individual stills, Wonder Camera owners would simply have their pick of perfectly crisp photos as frames grabbed from video."

Maybe, but there would have to be a way to have it integrate over several frames to get long-shutter-speed effects, and there would have to be a way to "tag" points in time so you're not sifting through hours of footage later to get the stills you wanted. Much of what this article proposes sounds like it just defers work (choosing aperture, choosing shutter speed, choosing point in time) to a later point, something I would NOT think photographers would want to do, since they already complain about how long post takes!

4) Finally, even this concept turns out to be 100% correct, I don't see it changing the art of photography all that much. As other people here mention, good photography is about good subject matter, framing the shot, clever use of lighting (whether artificial or available), post-processing to get the desired artistic effect, and being in the right place at the right time. Technology can certainly assist us with these things, but fundamentally it is these human element, not the technology, that makes photography what it is, and that won't change. As technology improves the photographer's decision process focuses less on making technical trade-offs (like shutter-speed vs aperture, high-ISO vs noise), but the creative aspects remain as important as ever.

+1 for the Kinesis, great keyboard, fixed my wrist pain!

Usually at some point I start yelling at myself "ok, you REALLY need to work now" and once I get into it I'm extremely focused and productive, especially if a deadline is looming.

My typical day:

9:30: wake up

10:30: get to work (now my home office), check email

10:30-noon: procrastinate

noon-1:15: lunch then 45 minute walk

1:15-5:00: procrastinate

5-7pm: 2 hours of amazing productivity

7-8:30: Cook & eat dinner

8:30-midnight: Take rest of night off or perhaps have another couple hours of productivity.

It's not as bad as it sounds cause usually the "procrastination" is me experimenting with new code or programming languages (or reading HN), just not what I'm "supposed" to be doing.

My advice: Get really good at C in linux. Focus on learning about cool algorithms and datastructures (hash tables, rb-trees, alpha-beta search, shift-reduce, ...) rather than the particulars of any language. If you get great at C programming in linux, and have a solid algorithmic foundation, then you'll be able to pick up the language-du-jour in a matter of days (if not hours).

holy cow! hey Brad... it's funny who you run into on hn!

I think design and profit margins go hand-in-hand. If you have a low margin business selling something cheap in large volume then design doesn't matter so much. If you want to sell higher margin, boutique, premium goods then design matters more (hence the word "designer").

Good luck to you! It's inspiring to hear about starting a company on pocket change!

I'm wondering if you're allowed to use those "New York Times", CNN, etc logos... sometimes companies can be protective of their branding.

Do you plan to form an LLC or sole proprietorship?

Sadly, I feel the same way about "organic" and "free range" labels. I wish I could feel good about spending more money to buy sustainable food, but I think it's mostly bogus marketing.

Does this mean that by pushing the deadline earlier they: - Didn't get enough applications? - Didn't get enough quality applications?

Or am I reading too much into this?

What's beautiful about them? Many of them look overly complicated.

"Tax the Rich" seems to be a popular sentiment (probably because the majority of people aren't rich and would rather anybody but themselves pay more in taxes), but is it effective? The richest 1% of the population already pays for 40% of the US's total Federal Income Tax revenue, and the top 5% pays for over 60%. (http://www.ntu.org/main/page.php?PageID=6). And because of the Laffer Curve (http://en.wikipedia.org/wiki/Laffer_curve), increasing taxes to the rich could very likely result in LESS revenue.

I hate it when I spend money on software just to be treated like a villain.

What happens when the license server gets taken down? I still play 27-year-old NES games on occasion. Will Ubisoft even be around in 27 years?

I see...we need Universal Algebra because Abstract Algebra isn't abstract enough.

I'd put it this way: If you find yourself spending significant time inside a debugger, you should probably rethink how you (or your teammates) have been developing code. Time is certainly better spent preventing bugs with unit tests, asserts, thorough documentation & thoughtful design.

That said, on the occasion when a bug does crop up (which will hopefully be rarely), the debugger is usually the fastest way to find the problem. A good programmer will be sure to root cause the problem rather than fixing the symptom. Debuggers are indispensable tools for working up the causality chain.

I know it's high time I learn how to use git, but perforce is what we use at work and their 2-user/5-client evaluation version has served me well at home. Just letting those perforce users among us know that it might be worth upgrading. I know shelving has been on my wishlist for a while!

Malkavich Malkavich, Malkavich Malkavich Malkavich

What software do people actually use to capture the screencast? Is there anything good for linux?

I find "Don't reinvent the wheel" to be a terrible idiom. Think about the wheel for a moment. Is the wheel on your bike the same as the wheel on your car, or on a train, or an an airplane? Of course they're all different! The wheel has been reinvented thousands of times, and we're better off for it!

Yes, it can be worthwhile to borrow successful concepts from other disciplines. But in many cases, the concrete implementation NEEDS to be reinvented in order to apply the abstract concept to a new purpose.

>> The vast majority of students are there to get the piece of paper...

And, in BYU's case, to meet other Mormons.

This guy completely misses what computer science is about. Sure, nobody studies "Toaster Science" just like nobody will ever study "iPad Science". But plenty of people study mechanics, electrical engineering, thermodynamics and other fundamental academic fields that have enabled humans to design and mass-produce toasters.

At it's core, Computer Science is the mathematical study of computation and algorithms. Some of the most important results in CS were discovered before computers even existed (like the Church Turing Thesis). A new device with a slick form-factor and usable interface will not at all make CS obsolete.

If it could efficiently solve NP problems...THEN perhaps some computer scientists would out of work ;)

This diagonal method is also how you can prove the halting problem is unsolvable.