Well, localstorage is still much better than cookies. Browser includes cookies on all requests to your server. There is no way javascript can tell browser to stop that!!! Now, stop using cookies!!!!
with the CSS Paint API you instantiate a new canvas like object for every time your reference the painter, while this was referencing one specific instance of a canvas, right?
Yes, the only downside is that you have to procedurally instantiate a canvas. You get to control when the render event should happen though. With CSS Paint API, it feels like declaring webgl shader.
I am afraid once a successful class action lawsuit is unleashed, many similar lawsuits will be in the rise. It will definitely stifle innovation, defeating the very purpose of kickstarter
To be fair, $200 is really not a lot of money. For all the contents, videos, interviews, samples, one can imagine how much effort being allocated for this. Being famous doesn't entitle someone to be able to charge expensively. Likewise, nothing is stopping anyone out there ("nobody" or not) to charge the price he or she wanted to. This is what we call internet, it's completely free market. No government or regulatory body is going to say "Hey Nathan! You are nobody, stop charging people a bloody $200 notes for a piece advice." Let's put some angles of perspective in, if he successfully sells 1000 copies at $200. He will make $200k revenue. Deduct that with the cost of his labor, his lost opportunity to work on his apps, marketing fee and the support time cost. He will probably pocket 30-40k pure profit. What if he has to follow what people say to price it lower at $20. Selling 1000 copies will barely cover his cost. And where are you going to find 1000 people who will believe this $20 crap going to give you top notch advice on delivering amazing web apps UX and design. Charge premium is a good start. Keep improving the content is a sure way to gain you the loyalty of customers.
Back to the odd marketing strategy. Why show it here? Why HN, really? Most will simply want something quick and simple to show the idea behind their hacks to the community. We are frequently being shown unfurnished prototype (indeed twitter bootstrap half the time). Anyone really serious about building well designed and engineered UX web applications will probably have a well funded startup with specialist UX designer onboard. Or someone must have been a long time practitioner of UX to be bothered about putting much focus on UX. It's not as simple as reading a book, listen to an interview, you become great in UX. It's only through years of experience, going through iteration after iteration of UX design tweaks, you will have a basic idea where to lead the good UX direction. A long the way, there are still huge chances to make UX mistakes here and there. Too simplified, or too complicated. Too plain, lack of creative input. Too much boring UX treating every users as dummy as they can. I am sure it will be hard for the community here to start bothering UX before getting their hacks done. And worst, the next hacks are waiting ....
The catch: don't go for native scrolling if you want iOS issue fixed. Try emulated scrolling what iScroll did before iOS5 came out. When inertia scrolling kicks in (the moment your release your touch and page continue its momentum scroll), any offset, scrollTop information is not updated. You are out of luck trying to fit any positioning calculation based on outdated offset/position. At the moment, there is no solution for this until Apple fix it.
I haven't really try LZW compression for the image data, but it seems that generated PSD file for a single webpage is extremely huge (multi megabytes PSD file). Standard Photoshop App generate image data with mostly Packbit compression. There is probably room to improve on that.
It will be so much better to have Objective C Runtime on Android. I haven't seen any java based Android apps that run smoother than Google Chrome on Android, which I highly suspect that it was built using C/C++ instead of Java.
totally agree. Android "Browser" is like the IE6 of mobile web. I bump into so many problems trying to tame android "Browser". Lack of hardware acceleration for simple displacement animation, the best transition animation you can apply is probably fade in and out not more than 2 stacked layers. The canvas was missing toDataURL method, canvas pixel manipulation with alpha channel resulted in some fuzzy pixels scramble. Weird virtual keyboard behavior. It takes few seconds for the device to tell the browser that orientation has changed, sometime has to be accompanied with light shaking of the device. No svg support for some older models. And majority of android devices out there are not running the latest Android OS. Luckily Chrome is much better browser than Android "Browser". Kudos to chrome team to make the mobile web on Android awesome again.
