Linux
Setting Up the Canon iP1700 printer for use on Fedora Linux
After much trials and tribunals, I finally got my Canon iP1700 to work with Fedora 11. Canon doesn't provide Linux drivers for this particular model so it wont work out of the box, but it is more than willing to print along happily with borrowed drivers.
I used the official Canon drivers for the iP2200. You can grab them from here: http://software.canon-europe.com/products/0010231.asp. In the OS selection pick “Linux”, choose your language and hit submit. Download the "Canon Print Filter for Linux (2.60 rev 3)" (or newer) package. Its a 22MB “.tgz” file.
Double click it to bring up “File Roller” (or use the terminal if you are feeling frisky). Extract the file “ip2200_Linux_260.tag.gz” to the Desktop. Extract the “guideip2200_2.60-1.tar.gz” if you want a manual. (Yeah, right.)
You will need to re-extract “ip2200_Linux_260.tag.gz” to a sub-folder on the Desktop (I call it ip2200), since it is also a compressed file. It contains these RPMs:
Go to the sub-folder ip2200 on the Desktop, double click on “cnijfilter-common-2.60-1.i386.rpm” to install it. You may have to supply the administrator password and choose “Force Install”. Do the same with “cnijfilter-ip2200-2.60-1.i386.rpm”. This should install all the necessary drivers.
Note: If you have attempted to install these drivers before (and have been unsuccessful), the packages will conflict and you will need to completely remove them. The command is:
su -c 'yum erase [name-of-package]'
For example:
su -c 'yum erase cnijfilter-common-2.60*'
Once the drivers are installed, restart CUPS by typing the following at the terminal:
/etc/init.d/cups restart.
Now to setup the printer through the CUPS server.
- From you browser go to this URL: http://localhost:631
- Choose the “Administration” Menu
- You may be prompted for a password. The username is usually “root” and the password is you system's root password.
- Hit the “Add Printer” button and select “Canon iP1700 (Canon iP1700)” from the list.
- Add a name, description and location or simply accept the defaults.
- In the model selection list hunt for “Canon iP2200 Ver.2.60”
- Choose a few configuration options and you are done!
At this point you may want to print a test page. You can do it through CUPS. To do this go to the “Printers” section, click on “Canon ip1700” (or what ever you called it) and from the drop down menu choose “Print a Test page”.
If you found this page helpful, do consider sharing with others who may be looking for something similar. Digg, twitter and links on forums work great. Let me know if the method worked for worked for you. (or perhaps didn't!) Or if you find an error in what's described above. Use the comment form below, or mail me.
How to Record a Screencast on Linux
su -c 'yum install istanbul -y'
Or you can use the package manager [System » Administration » Add/Remove Software]. Ubuntu users might want to refer to the community documentation.
Fig: The Package Manager
Once installed, run it. It shows up as a red dot in the task-bar. Right clicking it brings up the options and configuration menu, and left clicking it starts the recording. You can select an area or window to record, or record your whole screen. Some basic scaling options are also available.
Fig: Taskbar Icon
Fig: Options Menu
To stop recording, click the icon again. The application will then prompt you to save the screencast. Istanbul currently saves to Ogg Theora format only. The saved video can be played back on VLC media player.
Fig: Save Options
Sharing the Screencast
Dailymotion is the only major video sharing site that supports direct Ogg Theora upload and sharing. Youtube and the others transcode it to propitiatory formats. But that should hardly matter, since the end result achieved either way. There is also a HTML < video > tag. Ogg Therora video is supported via the Video tag in Firefox and Opera. Here is a short sample screen cast in which I use Linear's SPICE simulator to demonstrate a series positive clipper circuit.Phantom of The Opera
After months of exciting users with its new astounding new features via an alpha, beta and RC, the final build of Opera 10, the most awaited browser on earth, has finally been released. Opera 10 makes leaps and bounds in browser innovation. It add new and unheard of features and improves upon some old ones.
Opera 10 was meant to release with two major innovations. One is the "Turbo" feature for slow speed Internet connections. It works by connecting to a proxy server managed by Opera Software, and transferring cached and compressed versions of the requested web page to the user's terminal. The system is smoothly integrated into the browser and switches over automatically if Opera detects you are on a slow or weak connection.
Here is a humorous video from Opera, explaining the compressing technology: http://www.youtube.com/watch?v=tYEUFwJXVvw
The other feature is "Unite", which was meant to allow user to user file-sharing. It did not make it to Opera 10 yet, it is still under development. Apart from the major innovations, the UI has been totally overhauled. The new UI is much smoother and aesthetic. Many more subtle changes have been made to the browser. All of which work to enhance your browsing experience.
Opera 10 is also one of the two major browsers, as of now, to pass the Acid 3 test. Safari being the other.
The improvements to the development tool, Dragonfly was a tad bit disappointing though. However some improvement has been made. Web-developers however might want to hold onto Firebug for a little bit longer.
Opera Linux users will appreciate the much smoother Qt UI. The browser look and acts much more native to the system. Here is a screen-shot of Opera 10 with the Bookmarks side-tab open on my Fedora Desktop:

In other news, the logo has changed as well!
Opera 10 brings new and innovative features, looks and feel and a whole new level of browsing experience to the table. In my books, its the best browser out there. How did you like the new Opera?
Links:
- Opera Homepage: http://www.opera.com/
- Community Launch: http://bit.ly/2OylE7
- What's Missing: http://bit.ly/BiDGa
Embedding fonts in PDF files exported from Openoffice Writer
Over the course of your life on Linux (or Openoffice), you will come to a point where you will face a daunting task. You will one day find the inadvertent need to send some of you work, your hard-typed work, which you so happily created using you open-source tools, and which you so lovingly adorned with open-source fonts, over to a someone who more than contended to pay Mr. Gates hundreds of dollars every couple of years. In short, you will one day need to send an OpenOffice Writer document, laced with open-source fonts over to someone who uses Microsoft Office and has never even heard of Bitstream Charter or DejaVu Sans.
Windows users note: OpenOffice is also available as a windows download. The steps described here are identical for the windows operating system as well. "PDF export" is enabled by default in the windows package.
The easiest way to do this would be to export the document as a PDF. You will find the option under the menu File > Export to PDF. If you cant find this option under your install of OpenOffice, then you need to install the additional PDF export plug-in. Under Fedora the command is:
su -c 'yum install openoffice.org-extendedPDF'
Graphically, this package:

But the problem of the fonts still remains. By default, during the export the fonts will not be embedded into the PDF and the best possible substitutions will be made when the PDF is opened on the remote machine. To embed the fonts and ensure that the documents looks the same on all platforms you need to select the PDF/A1 option in the export dialogue.
This one (Highlighted with a Blue Box):

Note: This step is identical on Windows.
To check if the embedding has worked, you can use the pdffonts command. This script checks the PDF files and prints embedded font information. The usage for the script is:
pdffonts Filename.pdf
The script accepts parameters to limit the page range and enter passwords. Use the -? argument the see the usages.
The script in action, checking a PDF file 'report.pdf':

emb stands for 'embedded' and sub stands for substitutable.
Refreshing PackageKit
PackageKit is the default package management system on Fedora from versions 9, 10, 11 and up. Package kit has great many advantages, read about them here. But for all its merits, it has one tiny niggle. Sometimes, especially on new OS installs, or after adding software sources, it fails to automatically refresh its cache. And the GUI (currently) has no way of doing that for you. So you will have to show it some good old command line love.
When PackageKit fails to refresh the cache, you either get an error that reads "No results were found. Try entering a package name in the search bar." for every category of packages, or the packages you expect to show up, from the newly added software sources, don't show up.
Graphically:

Refreshing the cache from the command line is simple. The text mode program that allows you to talk to PackageKit via the command line is "pkcon". To refresh the cache type this at the prompt:
$ pkcon refresh
pkcon will run through the refresh routine and PackageKit should work as expected.
Graphically:

If you wish the explore pkcon a little more, type this:
$ pkcon -?
into the command prompt to print the pkcon help.
You might also like to learn that pkcon can not be run as a root user. GTK+ applications should not be run as root. Doing that blows huge security holes into your system. Read more here. If you are in the mood for a little more reading you might want to look into this interview on fedoraproject.org. It talks about the rationale behind the inclusion of Packagekit into Fedora, its advantages and future goals.



