Unity3D gets down with the native client!

http://blogs.unity3d.com/2011/06/02/...er-plugin-api/

This is a breakdown of Native Client (NaCl), and a probe into the poorly documented Pepper API to find out what system features NaCl will have access to.

What's most encouraging is that Unity will be providing a platform target for its upcoming versions of Unity! That means, it will be very easy to port Unity developed games to the web using NaCl.

Bottom line? Get ready for this....

...in the browser. Zero install. Play anywhere.

Linux fans should REJOICE that there's finally an alternative to installing windows on their machines for playing modern games. While NaCl will be a large part of Chrome OS, it will also be built into Chrome and likely Firefox and Opera as well.

Bring on the revolution.

It seems that there's a war a-brewin'

Mozilla is flat out rejecting inclusion of Native Client in its browsers!
http://www.theregister.co.uk/2010/06...an_on_firefox/

Now, I can understand the rational, they are championing the open web and hope to fill NaCl's role with Javascript, but it's very short sighted aim.

Javascript has some serious performance limitations when compared to native code. Now, javascript is improving all the time, but there is a tremendous disadvantage that javascript apps have on even the fastest engine.

NaCl can easily become an Open Standard in that it is open source, offers serious benefits, and can be implemented as a plugin.

What Mozilla is effectively doing is disallowing web citizens to vote with their feet, and rather pushing the decisions of the future of the web into small, closed committees that determine the eventual 'standards'. By taking such a staunch side with this technology, they undermine the 'open' characteristic of the web. I also wonder if the updated PPAPI that NaCl will rely upon will be implemented in the upcoming Firefox to accept NaCl as a plugin.

I would have much preferred if they hadn't drawn their line in the sand so rebelliously, but rather conveyed that they were keeping an open mind despite respecting Open Standards.

I think that it's also very short sighted. NaCl effectively eliminates Flash and other proprietary plugins, by removing the very need for a plugin. In this way, it will actually reduce 'fragmentation' on the web pushing all code into a similar bucket.

I think that NaCl will disrupt the web-world, and bring desktop quality applications to the browser. Mozilla users will lose out on these experiences, and I predict that their market share will continue to lose out to Chrome and players that adopt these technologies.

I forsee that it will form a fissure down the center of the web -- those that adhere tightly to standards, and those that are willing to experiment with newer technologies.

I also foresee that a standards body will extend ECMAscript or create another statically typed language that allows for fast JIT code to compete with NaCl. I would be cool with this as well, just so long as there's an intermediary bytecode to compile to, to allow for choice in language. Interestingly, even with this solution, NaCl would still be a suitable umbrella.

One way or another, native code is coming to the browser, which will usher in a new era of games, simulation, content-creation and productivity applications. The question is -- how late to the party will Mozilla be?

Here's a bloggers response to Mozilla's rejection of NaCl:

Native Client Hurts the Open Web
http://chadaustin.me/2011/01/mozilla...-the-open-web/

Here's an early paper on PNaCl, the portable NaCl implemetation:
http://nativeclient.googlecode.com/s...site/pnacl.pdf

Oooh... I love the idea of platform independent bitcode compiled to native running at near natively-compiled speeds (possibly native speeds with adequate optimization). This gives NaCl a truly device independent format for efficient execution on a variety of platforms. Any language could conceivably compile to this intermediary format!

This is a step in the right direction.

More information on the LLVM project can be found here:
http://llvm.org/

For those interested in taking a gander at the LLVM instruction set take a look below:
http://llvm.org/docs/LangRef.html

A little reaserch on LLVM has revealed that among the front ends available is on that uses Java bytecode and compiles to LLVM bitcode. It is thus possible for PNaCl in the future to support applications written in Java without requiring a separate interpreter. Additionally there are front ends for many languages including C/C++.

This is a fantastic target for universal code. PNaCl is not native, but with aggressive optimization it can run with the near efficiency of native code, but in a language, operating system, and architechture independent way.

I'm surprised LLVM isn't at the center of more OSS projects.

Amazon, known for their online shopping experiences, is probably less known for its cloud services.

Recently the company held a technology open-house giving a tour of their impressive data-centres.

http://www.datacenterknowledge.com/a...-data-centers/
PRESENTATION PDF: http://mvdirona.com/jrh/TalksAndPape...se20110607.pdf

It seems that Chrome almost replaced Firefox in Ubuntu! It also stands a good chance of making it into the next release.

http://linux.slashdot.org/story/11/0...d-FF-In-Ubuntu

I would have *much* preferred if it had replaced FF. Chrome, IMO, is a far superior browsing experience, and a joy to use -- plus, I'm very pleased with the aggressiveness of Google in optimizations and new features.

On my netbook, chrome had transformed the experience from being slow and confined to a quick minimalist browsing experience in Ubuntu. It makes the experience quite fun, rather than feeling like a compromised experience. I would like to see how it compares to the latest FF.

+1 for Chrome browser from as well. I run Chrome canary build in WIn 7 and Chromium 14 on arch linux, both perform better than their Firefox counterparts.

As I don't develop, my need for the cutting edge devel tools is a non-issue, and can't really say which platform is better for a developer. But for sure there is a major push from the community to get Chrome ahead of FF.

I love how snappy Chrome renders pages, the url-bar-less screen real eastate (alas, not an option apparently on Chromium, at least on the current Arch Linux build), the Chrome webstore etc.

Also, I find the extension "buttons" better implemented on Chrome again saving screen space, as opposed to FF's add-on bar on the bottom - possible to tweak that one I guess, didn't research enough - after all everyone claims FF is the most customizable of all browsers...

I ran sunspider benchmarks on both Win and Arch for canary/dev build Chrome vs FF nightly (FF-pgo on Arch), in both cases if memory serves correctly, Chrome actually came ahead.

My nod for the last 3 years has been for Chrome/ium.

Google is once again innovating on top of chrome. This time? Page prerendering:

http://blog.chromium.org/2011/06/pre...in-chrome.html
VIDEO: http://www.youtube.com/watch?v=_Jn93FDx9oI

This technology attempts to fetch a page and its assets that you're most likely to click in order to make page loading instantaneous upon click. Width inexpensive bandwidth and faster internet speeds, this technology is certainly welcome for a faster web experience.

It is being implemented with google.com, though is baked into the dev version of chrome for future releases.

Very cool.

Samsung Series 5 chromebook now shipping:

http://www.engadget.com/2011/06/15/s...ic-white-tita/

This is a very good looking system.

EDIT: The reviews are coming in, and it's looking positive:
http://www.amazon.com/Samsung-3G-Chr...8155744&sr=8-4

I have my own opinions of the Chrome web store, but I will save them for another post.