Would be awesome to see some of these games implemented with Node.js and HTML5... Angry Birds and others have demonstrated that HTML5 is good enough for graphics, and Node.js is a good fit for server-side logic.
I'll argue that not everything about this is "sad". Knowing JavaScript opens new opportunities when it comes to the kinds of products you can develop and kinds of audiences you can reach. You don't have to look at it as a burden. JS might not have the best design, but it's not horrible, either. It's just a tool, and it's up to you how to use it.
I agree with this and the Article, I was a front-end guy who became a good back-end guy, who spotted the opportunity early on for serious development work to be done back on the front end. There are not a lot of really good language guys over on the front end as the article implies, but I think that will change with time.
Just 2 years ago the whole thing was in flux, was Flash or Silverlight going to take the interactive web crown? Was a backend focused toolkit like GWT going to save the day and keep the developers on the back end? I would say that the score was not settled until as late as mid last year.
Around the time that jQuery was crowned king. Their is no doubt now where web development is heading, disconnected browser apps communicating to the server via REST and JSON. (can I take a moment to tell all the haters, I told you so). All kidding aside, now it just remains to be decided what the traditional pieces of that client stack are.
Dojo offers a compelling and robust toolkit, that I personally like very much and recommend to people doing large apps in the browser. Then there is jQuery, the juggernaut, with something like 40% of the mindshare it hard to not declare jQuery the winner and move on.
The problem is for large application jQuery on it's own falls down miserably. It has to be paired with the likes of backbone.js and underscore.js to provide any basis for large app development and even then it is still a little more spartan to develop with the jQuery/backbone stack than it is Dojo.
While it is very easy to just call the game for jQuery, Dojo provide a scrappy underdog and a dark horse in this race. The point is, once "the one" toolkit is decided I think you will see more developer migrate over, and I think jQuery's dominance is starting to become that transition point. Making them feel comfortable about what they are investing their learning time in. This is an area where developers cant afford too many missteps and false starts in, if one wants to remain marketable they must choose the technologies that offer the greatest prospects, until recently that was not very clear on the client side.
For me, as of now, I keep one foot on each side of the fence until the dominate framework of the client side emerges. I think Dojo still has a long shot in this as people come over and realize that Dojo is the most robust toolkit available.
Sarbox adds additional accounting procedures, with the goal of preventing another Enron or Worldcom, where lax accounting procedures meant that they were telling their investors "everything is fine!" one day and "we're bankrupt, your stock is worthless", the next.
The contention is that these additional procedures disincenvitize companies from going public. My counter-contention is that they're a very small drop in the bucket compared to the pre-existing incentives and disincentives.
"Increasing the value of the firm they are investing in? Great job guys."
Did Goldman increase the number of Facebook's users or add a new awesome feature? Did they address the privacy concerns of Facebook's users? Did they make Facebook experience better for the users? No, no and no.
All they did is increase the "perceived financial value" of the company, which as we know from not too long ago, is intangible and fleeting at best, and total BS at worst (Reminds me of the $100bn "valuation" that 37signals got http://37signals.com/svn/posts/1941-press-release-37signals-...)
As far as I am concerned, Goldman can do whatever it wants as long as its actions don't have negative effects on the rest of us. Internet bubble of 90's hurt many people, and so did mortgage bubble of late 2000s. I don't want Goldman to contribute to and profit from another one of those. How do we make sure it doesn't happen?
No, because GS isn't an engineering firm; they're a finance firm, so what they do is decrease the cost of capital for Facebook's own investments, which might include:
* Rolling up other companies in their space (Yelp, Twitter, 4square, Instagram, Rdio, who knows)
* Buying their way into a strategic but lateral market (search, advertising, mobile, hardware, ISP, who knows)
* Drastically improving hiring and retention by improving comp packages and adding headcount
* &c &c &c
Your comment seems basically nonsensical, suggesting as it does that there is no way to improve Facebook that isn't visible in a commit log.
I also object to the comparison between the Internet bubble and the mortgage bubble; the structural causes of both were different, the underlying value of the tulips being sold in both were different (at the end of the Internet bubble we had, you know, an Internet), and most importantly the mortgage bubble was a financial black hole that sucked every homeowner involuntarily into its gravitational field, whether or not they were jackass speculators.
Another impressive aspect is how quickly Kinect hacks are coming out. Some products seem to strike a chord with the enthusiasts, plus all of these things look very cool in videos, which makes for good publicity.
I think it takes a mixture of product quality vs hackability vs popularity. There's little point in being huge in any if you don't have all three.
Kinect is insanely popular - by its sales figures. It appears exceptionally hackable - given how quickly and how many hacks have come from it. Finally it appears to be of exceptional quality (at least for the $ value) for these hacks to actually be using it so innovatively.
"To work around that problem hospitals needed to, as one medical physicist put it, essentially trick the machine into thinking it was using a different attachment, which it did recognize. To do that, users had to enter additional data into the SRS system."
Very. Presumably they're paying for hefty support contracts as part of the purchase of these machines - I cannot imagine that the developers recommended medical end-users alter the system. If so they seem culpable.
