triplefox

I've met and spoken with some of the "instant successes" in Flash gaming. DTD, Filler, etc. From the little bit of discussion I had, I got the impression that they didn't really have any profound understanding of how or why the game they made became hugely popular. Mostly, they were benefiting from right-place-right-time effects, from being relative outsiders to gaming, and from having just enough skills to do the job.

Basically, trying to crank out games quickly isn't really a recipe for success. It is a good starting place for creative ideas, though. You can start a project with one thing in mind, build it up to a certain stage, leave it a while to think about, and later come back and iterate on the whole concept.

But if you are looking for reliable success in games, the time spent is way more important.

It took me a while to figure out what I wasn't seeing on these results pages.

SEO spam.

So I searched for "SEO spam" and it asked me to build a Square for it. No thanks >:(

Type systems have been getting better - more nuanced, more accurate/helpful static checking, though the mainstream is proving very, very slow to go away from the C model.

If you want Flex/JS interoperability: haXe.

This was written back in 1997 and is a timely read even today.

I'm going to speak in terms of techniques rather than tools. These are the best/fastest techniques I've found:

1. Procedural generation in your own code. People forget about this option because classical 2D games were too processing/memory-constrained to allow for anything but tiles loaded direct from ROM, but it's really great for environments. It works out to be very fast to produce because you can iterate and randomize and come up with a lot of variations without extra work, but you have to know what you're going for first. The download size savings is a bonus too.

2. Characters are cheapest to do with 3D only because of animation time. If you don't have a lot of anims, hand-drawn tends to give a more lively result, especially in low-res pixel art styles, but even then, painting over 3D will reduce your error rate. Invest in a Wacom if you want to get serious about drawing on computers, but start with the smallest size: you rarely need more.

3. When I need a clean cartoony look I drag out Inkscape and do a vector trace over the image I'm working with. This can take quite a while, but the results are really good when you go all-out and use two vectors per outline to give them variable width.

The default state of the world is free. Non-free happened because people decided that controlling content distribution for profit was a good idea, several hundred years ago. What if it isn't, or isn't anymore?

I can provide an example of how non-free would hurt you. If every site HN linked to was a pay-site then HN itself, Digg, Reddit, all those sites would suddenly cease to exist. All the fluidity of trading links back and forth would be gone as nobody would be able to afford to investigate what content is good anymore.

In fact, even Google would be in trouble if everything were a pay site. How could they possibly afford to pay to crawl every site on the internet? And, for that matter, what about all those startups that are based around data-mining existing web content?

The argument goes, "We can afford to produce better content by putting it behind a pay-wall." But when taken to the extreme, it's inefficient. No content is so good that it can replace the service of Google and other search engines. Or worse, in the ultra-non-free-world, Google would go from being a de-facto monopoly to a regulated monopoly; they strike up a pay-to-play deal for a site to even EXIST on their search engine and the entire Web is at their mercy. (we're not even talking about placement here)

It sounds ridiculous, but it's exactly the kind of thing we've let happen to copyright and patents in the past. By entering into these complex arrangements of enclosure, you end up with a battle between governments and big business that squashes the rights of individuals beneath regulation and fees.

This is an essay on revising civilization's economy in the face of abundant technology:

http://www.ascentofhumanity.com/chapter7-2.php

He starts from the thesis(previously established) that money earning interest is the source of our massive wealth inequality problem - interest ultimately favors a concentration of wealth on one person, and it also favors destructive practices that make a large sum of money immediately and earn interest exceeding the value of a sustainable solution.

From there he points to multiple alternatives that might bring us towards an equal economy without top-down planning.

The one that always bugged me but seems to be true for most dynamic languages is immutable types being passed by value and everything else passed by reference.

Also see: http://www.ascentofhumanity.com/chapter4-10.php

A similar-looking symptom I've seen with IE6 is that links won't "take" if I click and then move the cursor away. IE will display a message in the status bar when I click, but for some reason it is noticeably less likely to actually load the page if I don't keep the mouse on the link. This was particularly bad on dialup.

My dad knew someone who aimed to walk the full length of every street in San Francisco, at least one per day.

There are several thousand streets in the city, so it took a few years but the guy did complete his mission.

Hackerspaces now are what computer clubs were in the 1970's.

I think this more or less describes the life I'm perpetually trying(and only succeeding in brief snatches) to live.

I've been deep in the depths of hacking a new feature into an existing language - adding a reflective/referential data definition language on top of haXe. (It's called XReflect, if curious you can get it off haxelib)

Essentially, the problem I keep running into with referential data structures is that haXe, like most popular languages, will only let you set them up in an imperative fashion, one dependency at a time, which tends to both spread out and bloat application code. What I want is to write my dependencies in any order and let the resolution be sorted for me.

I'm pretty sure a language with AST access and quoting like Lisp makes it relatively easy to write something along these lines, but in haXe, I had to write about 500 dense lines of code, most of it going to parsing and fighting the limitations of the type system. It still has some major bugs that I'm trying to resolve with a refactor at the moment.

The funny thing is, I know I'm not just fighting haXe here, I'm also running into the limitations of the platforms that haXe compiles to. The newest strategy I'm applying to resolve references is to add a "__parent" field to the reference stubs, which refers to the container(an array or record type) of that reference. I use that later, after all references are resolved, to drop the final value into the place of the stub, but as it turns out this causes difficulty printing out the working structure. Neko will recognize the value of __parent as "....", but Flash 9/10 will recurse into the field again and get a stack overflow.

So I had to add more code to write a pretty-printer which is aware of such traps.

Ionic air purifiers emit a significant (potentially harmful) level of ozone. That might also be a problem with this.

Or if you want a strictly economic explanation: The law is a cosmetic blemish upon what could be a top-to-bottom free system.

The high level would be original free software and media - tools and inspiration.

The medium level is derivatives; things made using the tools and media built upon other media.

The low level is the consumer level, where the best things are filtered and distributed to the public.

The system takes the lowest-cost option at each step: remixing material and building on open source is easier than starting from scratch. Digital distribution is easier than physical distribution.

But it is entirely dependent upon the source material being permissive-use. When it isn't, derivatives become illegal material, and people will subsequently choose not to make value-added derivatives because any profit motive that would have existed is gone; settlement and legal fees are likely to wipe out any gains.

This doesn't affect the consumers of media, however, because consumers are largely anonymous and hard to target by the legal system. Thus you get widespread piracy, but few derivatives.

Unsalted nuts. They store really well. Walnuts in particular have plenty of calories but if you have too many in a day you'll be overloaded on omega-3, which will make your hands and feet swell painfully as pressure is applied; the blood vessels in them start breaking too readily.

So nuts aren't the bulk of my calories, but they help make everything else taste better.

It's like going back to the days when ROM BASIC was built into every "microcomputer." Only this time they're going for a full OS instead of just BASIC.

How dare they make money!!!!!!!!

If this view persists, the entire site's direction is going to change because some proportion will upvote only from "classic" and others only from "main."

On the other hand, I find that oftentimes I will try to discuss a programming topic in an online forum and people will say "Oh, just do x, that is the best practice." And then I will say "but it doesn't completely solve the problem, I'm trying to get this behavior as well, so the problem isn't so simple, it's just pushing the full solution into some ad-hoc mess later..." yet they will be unable to understand. It is disheartening.

I'd say it's almost 50/50. I'll go down more blind alleys, but I can also see more "cunning solutions" too.

It's an overly simplistic interpretation. The more poignant way to frame this would be a specialist-vs-generalist view; if you're overly specialized and never crawl out of your hole you miss opportunities for synergy.

Programming is a good "glue" for generalist approaches because it acts as a stepping stone to automatizing a specialized process.

My personal experience with this bears it out; an HFCS soda will be consumed quickly (and I feel unsatisfied) but a sugar one is better to sip through, and I get a better sense of its impact as well.

I discussed this yesterday (elsewhere) with respect to games. I couch it in terms of "permission economy" (the present strong IP situation) versus "remix economy" (the weak IP situation)

My analysis is that, first of all, there is a lot of waste overhead involved in large-budget productions. In a remix economy everyone will use the minimum budget possible to produce new material, so producers will operate near perfect-competitor efficiency.

Second, and perhaps more importantly, because all material can be reused, consumers will profit; new material will constantly seek to create new value, rather than to simply create a rebranded version of old value.

This can apply to drugs as well as media. Presently, pharma corporations each jealously guard their research. Hence they are all duplicating effort. In a remix economy, the company that kept secrets would be overtaken by collaborating competitors, because their costs would be too high. They'd get throttled from the overhead needed both to do original research and keep it secret.

Little-known fact: BART has been using 8-track technology for its ticketing since the system was created in the 1970's. You can see this when the kiosks need repairs.

This is why I've started adapting my desk to allow standing most of the time. I feel fresh even at 6 or 7pm. I tried standing all of the time for a day or two but I find that sometimes I want to sit too.

Indeed, the article doesn't really address anything. It basically concludes on the "sufficiently smart compiler" fallacy. Compilers only get better after you throw tons of engineering resources on them, and that situation only occurs after a language gains tremendous popularity for other reasons.

For example, both Python and Ruby have less-than-ideal implementations, and the avoidable flaws in their design are well known. Despite that, it's taken a long time to get to the point where anyone is even motivated to work on a reimplementation.

And on the other hand, Lisp has had scores of implementations, but it still can't reliably earn the prized "equal/better to C" benchmark scores. This if nothing else should inform us that performance considerations definitely do come into play at the language-design level.

OT: I was in Michigan to see my brother graduate with his PhD. Soon before we left we found out that he couldn't walk because of rules and regulations(he still had to defend his thesis)and so we didn't attend commencement, just visited, ate out, took photos, etc.

We went to the Henry Ford estate yesterday. He was definitely the hacker-entrepreneur type. A substantial part of his house was basically a early 20th century laboratory, with engineers working on secret projects.

For me the most important thing has been stability of time. If I can plan that on any given day, my "usual thing" will be to code a bit in the evening, that's exactly what I'll do - even if I miss a few times. But if I'm getting jerked around by external forces, be they errands, phone calls, travel, etc. then it won't happen.