pdf.css
Google


   


You are here: CodeIdol > Java > Java 2 Network Security > page: 30 31 32 33 34 35 36 37 38 39 40


The history of Java security has been parallel to the main releases of Java:
JDK 1.0, JDK 1.1 and Java 2 SDK, Standard Edition, V1.2

5
:
1. The JDK 1.0 security model was very strict. Local code was granted
access to all the system resources, while a remote applet was always
considered untrusted, and could be used almost only for cosmetic
functions, like the decoration of a Web page.

2. JDK 1.1 still considered local code as completely trusted, with full access
to all of the system resources. However, JDK 1.1 also offered Java
developers the possibility to apply a digital signature to the code they
wrote. By looking at the digital signature, the user on a client machine
could decide whether a particular remote code was to be considered
trusted or not. If untrusted, that remote code would run in a restricted
environment. If trusted, that code would be considered as a local code,
with full access to all of the system resources. The JDK 1.1 security model
was more attractive, but still presented several limitations. For example,
remote code with a trusted signature was granted full access to all the
system resources, as well as local code. So, even when you wanted to
grant a signed remote code, say, only read access to a particular file in a
particular directory of your system, you had to grant it full read access to
all your files and all your directories. Moreover, that code was
automatically granted the permission to write on your system, install other
code, open a socket, and a lot of other things. This happened without your
intervention or your awareness.

3. The Java 2 security model implements fine-grained access control. You
can now classify the Java code that is to run on your system basing your
judgement on the URL location where that code resides and/or the owners
of the code itself. The owners of the code are identified through their
digital signatures. Multiple signatures for a single piece of code are
allowed in Java 2. Possibilities are now endless. You can say that a piece
of code coming from a particular location and/or signed by particular
signers can only read that file and write in that directory. Other code
coming from another location and/or signed by other entities can open
only a specific socket, while still other code can be classified to have full
access. Moreover, in Java 2, even local code can be subjected to security
restrictions.

In this section we will demonstrate to you that Java is not a threat to security,
provided that your Java system is configured in the correct way. On the
contrary, the security features that are part of the Java programming
language itself can really improve the overall security of your system.

5
In this book, Java 2 SDK, Standard Edition, V1.2 (J2SDK) is sometimes referred to as
Java 2 SDK
You are here: CodeIdol > Java > Java 2 Network Security > page: 30 31 32 33 34 35 36 37 38 39 40


ADBRITE ads links
   
Related tags







Popular Categories
Unix books and guides

AJAX popular information
C# language guides
Windows books and cookbooks

.......








Business Key Top Sites

be number one
rate your site




    С 2009 года мы стали переводить структура сайта на различные языки. Сайт теперь будет содержать книги не только на английском языке, но также и на других европейских языках, в том числе и на Русском языке.

    Русский Polski Francais Deutsch
    support sitemap terms

© CodeIdol Labs, 2007 - 2009