Site Sponsors:
Oracle: Web Access to Java Applets 
There I was - happily updating to the latest version of Java under Windows when - boom - access to my own on-line applets suddenly stopped.

Since the red-faced security flaws of Java 6, it seems that our BROWSER Applets are still well beyond any Oracle developer's ability to protect an OPERATING SYSTEM?

The constructive ideas I have to offer range from shifting the safe-tech responsibility back to Oracle Developers to (1) keep the sandbox safe, or encouraging Oracle to (2) simply craft a safe-and-secure 'Lite Version' of Java for browser use... -Much as Sun did for JME, CLDC, and elsewhere.

Until the inspiration is received however, be aware that there is an alternative to exposing your personal information to yet another public register.

Work Around


Since the policy-prohibitions are coming from Oracle - not your operating system - on Windows all one need do is to search for "Java" under the "Start Program" (Search Programs and Files) tools, then select "Configure Java."



Using the above configuration tool, we can add things like the site URLs for http://www.soft9000.com, http://soft9000.com, or http://TheQuoteForToday.com to Oracles exceptions list:

(Oracle: Tool tip bundles (1.7 u 51) Missing ?)

In a like manner, feel free to take a moment to add your favorite sites to your Java implementation's "exceptional-site list-of-white":



(Both the Yoda-style rhyme, as well as the GUI pun, are indented!)


[ add comment ] ( 863 views )   |  permalink
TQFTD - Technology Review 
Perhaps the best thing about being technical is having a mission: A calling - if you like - for doing something constructive with our spare time.

My personal passion is for creating the technology behind what is required to collect, organize, and share huge amounts of data. Indeed, well before an opportunity to serve as a Principle Engineer at Informix (a major database company. To date we had the fastest database technology I know of), I was amassing a plethora of data (mostly quotations and recipes) to share.

Starting in 1987, the first technological trek spanned a decade of publication as "::::: The Quote For Today :::::." The robot-generated newsletter was posted in alt.quotations, as well as via a relatively short opt-in email list.

Early Technologies


To share the quotations we had to implement several RFCs. Developed in C/C++, the bindings for both SMTP and NNTP remain part of the STDNOJ / ABLE C/C++ Projects. Supported on every popular computing platform, the core was built using both Microsoft Visual Studio 6.x, and Borland C++ Bullder.

Web Technologies


At the time of this writing, the sharing of our collections is limited to what you see on TheQuoteForToday.com. While plans are afoot for resurrecting the old newsletter, the purpose of this post is to review the technologies others might find useful, as well.


Above is the official Open Source Project Layering. Aside from a Java <-> PHP encryption layer (not depicted), we have shared the major components on Sourceforge.net:

PHP CRUD: Designed to allow us to easily create, read, update, & delete indexed objects using PHP.

BANMAN: A system to both automatically, as well as manually, moderate access to web content. Designed specifically to stop content-plundering robots.

HTML Forms: HTML forms are used as the Web Service layer. The core of the B.E.S.T philosophy, HTTP forms allows for both automated, as well as manual, browser-based testing.

EasyHTTP: A Java demonstration of how to use web-forms to drive our web-services (BEST.)

Java / Java Applet: The web-based version of "Doctor Quote Online" on the site is a SWING GUI Applet / Application combination.

This web version of Doctor Quote uses EasyHTTP. An Android version of Doctor Quote is available, but not yet at any 'app store. (Maybe next week?)

HTML/HTML5/CSS/jQuery: All the latest research into all the latest ways of creating a better browser experience.

Unrestricted Passthru: Some works do not need to use BANMAN to protect their data. BANMAN monitors, yet allows the content scraping of these (several hundred) pages to remain unrestricted.

Indexing, Security & Searching Tech: ClassIO is how we prime, process, and provision our large NO-SQL data collections, as well as their subsets.

Not illustrated (and therefore not available for public re-use) is our unique keyword creation & search technologies. The excellent response times are cleanly demonstrated on TheQuoteForToday.com.

Like our Enigma-inspired "NO-HTTPS" security strategies (i.e. outside of a browser, isn't the very concept of 'standards-based security' a bit of an oxymoron?), our hyper-fast search tools & access routine must remain a prominent unshared technology. Why? Because customized, time-proven technologies such as those allow places like Google to be able to afford lots of little things ... like their own space program.



Thanks for the interest,

-Rn

[ add comment ] ( 1295 views )   |  permalink
Zombiewatch: Pre-Backup Safety Check 
A friend of mine (all-right - it WAS me :) has been increasingly disturbed by the amount of zero-length files he as been seeing.

While having a proper backup scheme goes the very long way toward mitigating the effect of data loss, one nevertheless needs to know when bad things are happening.

The Script


With all due respect to popular culture at the moment, I have ever called my empty-headed, zero-length files "zombies." -Not only can the zeb-files creep up on us unexpectedly, but tracking down their source might well just consume most of one's brain. (for example, having zero-length .proprieties files are not necessarily evil (etc.))


Ultimately - much like their popularized namesakes, file-zombies can also do some pretty strange things... -Especially to automated backup routines! (i.e. A file goes 'zombie, and is then propagated throughout a proper backup chain. --Better it is to inspect and delete any new zombies well BEFORE allowing them to over-write their adequately-archived namesakes?)


#! /bin/bash
rm omonitor.txt
mv monitor.txt omonitor.txt
zdate=`date`
zcount=`find /d_drive -type f -size 0 | wc -l`
zreport="$zdate found $zcount zombies"
echo $zcount > monitor.txt
echo $zreport >> monitor.log
zdif=`diff -q monitor.txt omonitor.txt`
if [ -n "$zdif" ]; then
tail monitor.log > report.txt
gedit report.txt
fi
echo $zdif

The way the Zombiewatch script works is simple: In addition to keeping a log, the script will monitor the difference between the last zombie-count and the present.


Once any change is detected, you will see the last 10 entries in the log file pop up.



Using The Shotgun


To use the above script, just change that d_drive place-holder to the folder where you keep your most important files.

Note also that the script complains about missing counter-files the first two (2) times it is run. (Think of those two reports as the well-oiled "click-click" of your new Zombie-watch shotgun ...?)

Booya.

Testing


Once primed, one can test the script at any time by simply changing to the script's folder, then updating the monitor.txt file.

e.g:

echo booya > monitor.txt
./*.sh

The zombie-report file will be both generated, as well as displayed.

Easy-peazy.

Where & When


At a minimum (I use a combination of cron, as well as the backup-account's .bashrc), be sure to run the Zombiewatch script BEFORE doing a backup.

Note that I also have a shareware title I have been working on (and off... and on... and off) for years now. --Let me know if you would like it to wing its way into the public domain, and we might just charge a few more neurons around it.

Infestations


p.s. For those who discover that they have an ACTIVE zombie problem, it might help to add a historical feature:
#! /bin/bash
rm omonitor.txt
mv monitor.txt omonitor.txt
zdate=`date`
find /d_drive -type f -size 0 > "$zdate.found"
zcount=`cat "$zdate.found" | wc -l`
zreport="$zdate found $zcount zombies"
echo $zcount > monitor.txt
echo $zreport >> monitor.log
zdif=`diff -q monitor.txt omonitor.txt`
if [ -n "$zdif" ]; then
tail monitor.log > report.txt
gedit report.txt
fi
echo $zdif

Using the above, whenever new zombies are detected, all one need do is a simple `diff` on *.found files so as to track 'em down to their source. ((See `man diff` for more information.))


May THIS source be with you... always.


-Rn



[ add comment ] ( 1471 views )   |  permalink

<<First <Back | 33 | 34 | 35 | 36 | 37 | 38 | 39 | 40 | 41 | 42 | Next> Last>>