Based entirely on HTML, Java, and JavaScript, the initial release of Netcaster had brief glimpses of promise between a constant variety of crashes.  This was under Win32, a platform that Communicator has been stable on for some time.  My first impression: No way Netcaster hits the ‘Net this year.

Netscape burned the midnight oil on the second preview release and finally was able to demonstrate some of the their key new technologies actually working:

1. Automatic Software Update: Acquiring and installing Netcaster was as easy as clicking on a link. The application is downloaded and installed with minimal user interaction using a Java based technology.
2. Capabilities Based Security: Netcaster will require access to your hard drive.  When it does, you’ll receive a dialog which asks what type of access if being requested, who is asking for it (in this case, Netcaster), and why you should trust this application (in the form of a digital certificate which gives you reasonable confidence that you should trust the requestor).  This type of security is vastly sophisticated and flexible than current offerings from Java 1.1. and totally eclipses anything currently offered by Microsoft.

Netcaster might well be the killer app for Netscape’s new LAYER tag in Communicator.  When launched, you’re presented with Netscape’s own channel along with a channel finder “frame” which allows you to explore other channels. As you start clicking away, selecting channels, watching layers zooming to and fro, remember, this is just JavaScript and HTML. It’s not a plug-in, it’s not Java, view the source if you don’t believe me.

The Netcaster product isn’t for everyone.  It wants to own your desktop as many channels are designed for full screen mode.  This is apt to annoy anyone who is doing actual work on their computer.  At the very least, Netcaster is an impressive demonstration of possible directions for web pages, it’s a direction were web sites finally rid themselves of the static page format based on a book metaphor that never made sense for a computer in the first place.
 

[rant]  Security over the Internet got a slap in the face when a group of individuals successfully decoded a message encoded using the Data Encryption Standard, DES.  The message they decoded read “Strong cryptography makes the world a safer place “.

A similar effort is ongoing to crack a much stronger algorithm, RC5, using a rag tag coalition of hardware spread across the Internet.  The point: Prove that RC5 can be cracked.  The real point: What’s the point?

It’s time to do the math.  The key to the first message that was decoded was found when 25% of the 72 quadrillion possible keys were checked.  As I said, RC5 is a much stronger encryption.  A current status report of the effort to crack the code indicates that 2.88% of 2^56 (72,057,594,037,927,936) possible keys have been checked.  At that rate, assuming that no new machines are added to the effort, it will take 2.6 years to check every key.  Chances are, they’ll find it much sooner than that.

At the Company, it’s very sheik to be PRO strong encryption.  It is a very simple matter to believe that the individual should have every right the strongest encryption possible.  It happens to be a bit of thrill to send encrypted e-mail using Netscape’s newest browser, but step back and consider the following:

Assume there is information encrypted on a computer.  Assume that the information could somehow save lives, but the individual who encoded the data would not agree to decode it.  What happens when a judge can no longer issue some type of warrant to force the release of the information?   Suddenly, encryption puts the individual above the law.

The governments plan hinges on the idea that it keeps a set of “keys” which allow it access to encrypted information.  Sounds like Big Brother doesn’t it?  Fact of the matter is that with very little provocation a judge in your town could issue a search warrant of your home, but chances are that will never happen.  See, the odds are you’re probably a decent human being and, short of brief solo visits to the commuter lane, you’re unlikely to present a significant legal threat.

The same logic applies to strong cryptography.  They only real use you’ll ever had for encryption is for sending your credit card over the Internet.  Strong and reliable encryption already exists to make that happen.

Clearly a point was made when the government’s DES code was cracked after only 5 months.  DES was implemented in 1972, well before anyone thought that Internet would link the worlds computer resources together.  While the ongoing RC5 cracking effort may prove that by linking hundreds of thousands of computers together, you can defeat the algorithm, what does it prove?  Yes, a key can be found.  And, yes, it took 2 years and 100,000 computers spread across the planet.

Sounds secure enough for me.