After that debacle, I wanted to buy a computer that just works with Ubuntu, out of the box. My requirements could be described thus: oversized netbook with at least two CPU cores, “virtualization assistance” CPU feature, no CD/DVD drive, at least a 768-line display, and someone’s assurance that it’d work with Linux. I wanted not widescreen (4:3 aspect ratio), but those displays have pretty much disappeared from the market lately.

Ubuntu’s web site states that the Dell Vostro V130 will work with Ubuntu 10.04 LTS. I wanted to install a more recent version, but that was good enough assurance for me to start with. I placed my order a couple of weeks ago and I got it in the mail this past Thursday. In this post, I’ll document my unboxing and first impressions running Ubuntu 10.04 on it.

It’s worth noting that the model I bought, certified for Ubuntu 10.04, had a surprising lack of options. Memory was limited to 2GB — no more, no less. The one hard drive available was 250GB, 5400RPM disk. The CPU is a Celeron Processor ULV U3600 (2M Cache, 1.2GHz, 800 MHz FSB). This configuration with Ubuntu 10.04 preloaded sells for $429 plus shipping as of today. More expensive models come with more memory, solid state or spinning disk, and a Core processor instead of a Celeron, and Windows — you can’t buy those models without a Windows license.

Unboxing

The computer came in a basic shipping box that was surprisingly light. Of course that’s because the computer itself is so lightweight.

Inside the box were the computer, the power supply with a three-prong mains end (extra adapter required if you want to plug it into common household extension cords), “Resource CD,” warranty, safety brochure, and dummy setup guide.

The overall appearance of the computer is exactly what I wanted: thin, non-flashy design. Built-in human interface components include:

• webcam
• stereo speakers
• two-button touchpad — apparently single-touch (see below)
• compact laptop keyboard with no number pad
• 13.3 inch 19:9 aspect ratio LCD display (1366×768 pixels)

The keyboard measures 11 inches from edge to edge, with the keys are spaced out 0.75 inches horizontally and 0.69 inches vertically; the keyboard includes a Super key (with a Windows logo on it). There is no pointing stick and no keyboard light.

The display has an anti-glare coating and is not glossy; it’s made for work, not media playback.

Front edge ports:

• 1/8″ mic input
• 1/8″ speaker/headphone output

Back panel ports:

• HDMI
• eSATA/USB
• ethernet
• VGA
• two more USB ports
• DC power input

Right edge port:

• SD / MMC / Memory Stick slot

Wireless connectivity:

• Wifi
• Bluetooth (Doesn’t work. More on that later.)

Altogether the computer weighs about 4.25 pounds (3.5 pounds if you don’t count the external power adapter). It fits nicely in my bag and doesn’t feel like a burden to carry around with me every day.

The battery is rated at 30Wh, and reportedly provides 2.5 to 3 hours of running time depending on who you ask. The system must be disassembled to change the battery. Warning: if you often need to work without access to AC power, the Vostro v130 is probably not a good match for you!

Observations on the Included Software

Since I’d rather be running a more up-to-date version of Ubuntu, I immediately set about making sure I had some method of going back to the off-the-shelf state and then testing all the important hardware features before erasing the hard drive and starting over. Short short version: Everything works except Bluetooth.

First, I had to start it up and go through the Ubuntu pre-installed system setup wizard. The most amusing part of this wizard was a click-through EULA from Dell that I was forced to agree to, even though the text of the EULA isn’t included in the wizard.

The message seems to indicate that the text is included in the Resource CD The EULA text is included in the “Warranty” booklet. I had a quick look at the CD on my office PC and it looked like a bunch of DOS-based diagnostic programs. I didn’t bother booting it or attempting to read the EULA.

Later on, the wizard offered to write a “Recovery” package to a fresh thumbdrive which would be made bootable. This package is intended to restore your hard drive to the initial state it had when it left the factory.

Hint: If you don’t want to sacrifice a thumbdrive to this process and just want an ISO image file, you can skip it here and then find the Dell Recovery utility in your Applications menu once you login. From there, it will allow you to write to an ISO image file as an alternative to a using up a whole thumbdrive. The ISO image is about 1GB. Store it somewhere else offline where you have 1GB to spare, and if you ever need it you can use the Ubuntu Startup Disk Creator (on any Ubuntu system or from an Ubuntu Live CD) to install it on a thumbdrive; or you can write it to a DVD and boot it from an external DVD drive.

In any case, you probably aren’t going to even need this backup image! The only difference I could find between the original configuration and a clean Ubuntu install was a Firefox toolbar. (Yay for minimal bloatware!) Be assured that you can always just download the standard Ubuntu 10.04 LTS installer and rebuild from that if you need to.

After I got to the end of the setup wizard I logged in and downloaded the latest updates for 10.04 and tested a bunch of things. Here are my results:

I knew about the Bluetooth failure before I completed my purchase. It’s documented at least once on the Ubuntu forums. Note that as of this writing, the product page for the Vostro v130 with Ubuntu states that the system includes Bluetooth connectivity. Technically this is poor or false advertising if the hardware is there but doesn’t work with the included software!

I’ll update this post if I find a fix.

Conclusion

I’m a happy customer and I got what I wanted. Over this past weekend, I was able to install Kubuntu 10.10 on it, and except for the Bluetooth failure, I’ve had no trouble with it at all.

If you’re looking for a cheap, thin, and light laptop to run Ubuntu, the Dell Vostro v130 is a good deal for $429 plus shipping.

Update (23 May 2011) I’m currently running Ubuntu 11.04 with KDE Plasma desktop and pretty much the same performance and compatibility results as shown above.

Update (3 Oct 2011) The Vostro V130 has been discontinued and it seems that its successor is the Vostro V131. I cannot find an option to buy this machine with Linux and not Windows, but the Ubuntu web site asserts that the Vostro V131 is Ubuntu-certified. I suggest that if you want one, call Dell on the phone and keep trying different ways to ask to buy the Linux version until someone is able to help you. If someone knows the correct store URL for this product please let me know!

I have about three and a half Linux hosts that I use on at least a semi-regular basis, depending on how you count:

• Thinkpad Tablet (keyboard, pointing stick, and stylus + touch screen)
• non-Tablet Thinkpad
• ancient (circa 2000) Thinkpad
• virtual machine that’s always running on my Windows 7 work PC , so I always have a Linux environment at work

Since virtually all apps in Linux store their per-user configuration information in text files and scripts somewhere under ~ (your home folder), it would be nice to share and synchronize that information across all my Linux hosts. For example, if I find the right Greasemonkey (user Javascript) script that solves a web site shortcoming, or write a shell script to get automate some task, I don’t want to be replicating that effort on each host. Nor should I have to manually synchronize all this stuff once I’ve configured it right in any one place.

For this project, I wanted to learn how to use the Bazaar version control system. I have much experience with Subversion at work and personally, keeping my office projects and personal notebooks in Subversion repositories, but I find Subversion takes more time than I’d like to perform a synchronization. Bazaar has some neato features that appeal even to people who are working on personal data and not collaborating, such as easier merging of parallel changes (compared to Subversion) and offline commits. Other next-generation distributed version control systems improve upon Subversion’s non-distributed architecture, but my casual research shows that Bazaar is one of the more newbie-friendly choices. Your mileage may vary.

Step 1: Initialize your repository

On the central SERVER:

mkdir ~/etc/config
cd ~/etc/config
bzr init
bzr checkout

Step 2: Move config files into into the repository

My concept is to keep all shared configuration files under ~/etc/config/PATH/FILE where PATH is the original path under ~ and FILE is the file name. This makes it easy to keep only the shared files in your Bazaar repository while ignoring all other config and data files.

mkdir ~/etc/config/etc
mv ~/etc/brendan.jpg ~/etc/config/etc/ # profile photo for social web sites
mv ~/etc/buddy-icon* ~/etc/config/etc/ # IM buddy icon (various formats)
mkdir ~/etc/config/bin
mv ~/bin/mount-* ~/etc/config/bin/
mv ~/bin/weather ~/etc/config/bin/
mv ~/bin/mailpda ~/etc/config/bin/
mkdir ~/etc/config/.config
mkdir ~/etc/config/.config/fish # Fish is my chosen command line shell
mv ~/.config/fish/config.fish ~/etc/config/.config/fish/
mkdir ~/etc/config/.config/fish/functions
mv ~/.config/fish/functions/* ~/etc/config/.config/functions/
mkdir ~/etc/config/.config/Terminal # XFCE Terminal
mv ~/etc/.config/Terminal/terminalrc ~/etc/config/.config/Terminal
# et cetera!

I hope you get the idea.

Step 3: Create a script to install symbolic links to all your shared config files

Of course you need to access these files in their original locations. Linux’s symbolic links come in really handy here. (Why does Windows still not have symlinks? :^( )

file: ~/etc/config/bin/install-shared-config

#!/usr/bin/fish
# install-shared-config
#
# This script installs links to ~/... from ~/etc/config/..., which is
# kept in version control.

# You'll need fish <http://www.fishshell.org/> to run this script
# -- or translate it into bash if you wish.

cd $HOME/etc/config
for i in (find . -not -path "./.bzr*" -type f -printf "%P\n")
   set d (dirname $i)
   echo $i
   mkdir --parents "$HOME/$d"
   ln --symbolic "$HOME/etc/config/$i" "$HOME/$i" 2>/dev/null
end

# make the script executable
chmod +x ~/etc/config/bin/install-shared-config

Then run the script and all your shared files will once again be available from their original locations.

Step 4: Check in your shared files

cd ~/etc/config
bzr add
bzr commit -m "Initial checkin"

Checking out the files on a remote host

Here’s how you add another host to the system — your laptop that you do most of your work on, for example:

mkdir ~/etc/config
cd ~/etc/config
bzr checkout bzr+ssh://USERNAME@SERVER/home/USERNAME/etc/config .

# Manually remove any already-local copies of config files found in
# the shared repository. The install script doesn't do this for you!

~/etc/config/bin/install-shared-config

Syncing a remote host

cd ~/etc/config

# Download upstream changes
bzr update

# What's new locally?
bzr stat

# Did you add new files since the last checkin? Run this to add all new files:
bzr add

# Did you rename files since your last checkin?...
bzr mv OLDNAME NEWNAME

# Did you delete files?...
bzr rm DEADFILE

# Done!
bzr commit -m "SUMMARY OF CHANGES"

Some notes about Opera web browser

Be careful what you put into your version control repository. For example, ~/.opera/opera6.ini would be problematic for at least two reasons: my username is embedded in it (my username is not consistent from one host to another), and transient stuff like window positions and recently used files are stored in it. I think it’s best to recreate my Opera configuration individually on each host.

The parts of Opera I do keep in version control are my user style sheet and my user javascript files.

I have a CSS file in ~/etc/user.css (which is a symlink to ~/etc/config/etc/user.css of course) that adds a thin yellow border around all links instead of whatever the site designer chose, so I can’t miss links. This file is installed using the Opera setting UserPrefs|LocalCSSFile.

And my collection of user Javascript files are in ~/etc/userscripts and installed using the Opera setting UserPrefs|UserJavaScriptFile.

Other applications you use might require similar specific tweaking.

More to come?

I’m just getting used to Bazaar for now, and my config files are a good low-risk data set to play around with. In the future, I will probably transition my Zim notebooks from Subversion to Bazaar and document it here. Using Bazaar should make it relatively easy to store an encrypted backup of my Notebooks repository on some public host, and keep it up today with a daily cron job; this way I will always have access to an up-to-date copy in case I’m on a host that is terribly out of sync — even if my main file server is offline!

I’ve had a Blackberry provided to me by my job on and off for several years. (Moving around in the company, I lost and regained the privilege/burden.) At some point, I decided I liked the Blackberry so much, I went to T-Mobile and got my own personal Blackberry that is not connected to my work. Why didn’t I simply use my company Blackberry for personal activity, when this is more or less allowed by policy due to the fact that it doesn’t incur any additional cost? The reason is that I like to keep a sort of personal firewall between work and personal data. I’m not comfortable using a company-owned device to access personal, private data like my email, contacts, and other web-based applications. I absolutely don’t want anything leaking from one side of the firewall into the other. Imagine this scenario:

1. I have an  incriminating photo of my cat doing something silly, stored on my company PDA.
2. I later capture a photo of notes on a whiteboard in a meeting and then attempted to email that to all the participants.
3. In the meeting email, I make an off-by-one error and attach the cat photo instead of the whiteboard photo.

Now this may be merely funny, but imagine if the wrong attachment wasn’t a cat photo but instead something I really would never share with work contacts? Keeping things fenced off in different PDAs doesn’t solve all permutations of this problem, of course, but it does take care of some of the worst cases.

I should also point out that this strategy doesn’t need to end with PDAs. Using separate hardware, or virtual machines, you can keep your desktop files fenced off too, and you should do that too. Just this morning, I sent this email:

From: Brendan
To: Kevin
Subject: Re: Vzaccess for Brendan all set

Thanks Kevin,

In case anyone’s interetsed, this if you’re using Linux (unsupported by Verizon) to connect via the Blackberry, use Berry4All < http://www.kuro5hin.org/prime-intellect/mopiall.html >. This might work in OS X as well. According to the Verizon download site, they don’t seem to offer support for connecting to Verizon wireless access from anything but Windows.

…

What the heck is “prime-intellect/mopiall.html”? It’s the URL for a novel. I was using my work virtual machine on my personal laptop last week to grab and convert the book for my Blackberry before I left for vacation. The URL for Berry4All didn’t make it onto the clipboard, and I didn’t catch this mistake before I sent it. Oops! Maybe if I need to be running Windows apps on my Linux laptop for non-work activity, I really should create a separate non-work Windows virtual machine.

Inevitably, data will spill across your work/personal firewall’s border. The best policy is to have your alternate environments ready to go before you need it, and in the moment, take a step back and think before you perform an action in the “wrong” environment. How risky is it? Did you double check file names, person names, attachments, etc.? Don’t do a lot of work outside the office unless you have additional computing environments.

If you only have a personal PDA and your company wants you available on the run, make it clear really need them to provide a separate PDA for that. Likewise, if you only have a work PDA, don’t use it to play around on Myspace; pay the cash to get your own, or go without it! If you have to work at home, ask for a separate company-owned computer or setup a virtual machine on your own computer. Don’t let your work files mingle with your porn pictures of your cats.

Where I will be staying, supposedly they have free Internet access, but you can’t always count on that. Just the same, I’d like to be able to bring some personal projects with me in case I feel inspired. If I do get motivated to work on these projects, I need access to tools and documentation that I normally get online

The following is a list of what I was able to find that may be useful:

XAMPP: Apache, PHP, Perl, and MySQL rolled into one package

XAMPP for Linux
XAMPP for Windows
XAMPP for OS X

If you’re accustomed to having your development files on a remote web server, the first thing you have to do to work offline is to get your server environment installed locally. If you’re using the LAMP stack, just download XAMPP from one of the links above and install it (offline) when you need it.

If you’re targeting some other web hosting platform, such as IIS or ColdFusion or J2EE, make sure you get that installed and working correctly instead!

Aptana: Client side IDE with good documentation built-in

I was scratching my head over downloadable documentation for HTML, CSS, and JavaScript. There are a few complete downloadable references for HTML and CSS to be found online (notably at w3c.org, where the HTML and CSS specs are maintained), but JavaScript is a problem. The closest thing I could find to a complete and free JavaScript reference was at the Mozilla Developer Center. Unfortunately, they don’t offer any way to download their documentation wiki.

A tip on Stack Overflow points out that the free Aptana IDE for web programming includes built-in offline documentation for HTML, CSS, and JavaScript, and the browser’s Document Object Model (DOM).

Here’s what to do:

1. Download Aptana and install it in your system.

2. While you’re still online, start up Aptana immediately and you should be prompted to ask if you’d like to install any additional plugins. Install the plugins for all the client-side technologies you’ll be using, even if you won’t be using Aptana as your editor. This way you’ll be sure to get offline copies fo the appropriate manuals and API references. Your choices for client-side technology plugins include:

• Aflax
• Adobe Spry
• Dojo
• Ext JS
• jQuery
• Microsoft ASP.NET Ajax Controls
• MochiKit
• Mootools
• OpenRico
• Prototype
• Scriptaculous
• Yahoo! User Interface (YUI)

3. (Optional) If you won’t be using the Aptana IDE to edit your source code, create a launcher script in your Aptana application folder to start up Eclipse (upon which Aptana is built) directly in the Aptana Help documentation:

launch-aptana-help.sh

#!/bin/bash

# folder containing Aptana
PROGDIR=`dirname "$0"`
cd "$PROGDIR"

# classpath argument for Java
CP=`find plugins/org.eclipse.help.base_*.jar`

if [ ! -f eclipse ]; then
 # Copy AptanaStudio eclipse launcher file to
 # 'eclipse' so that standalone.Help doesn't fail.
 cp AptanaStudio eclipse
fi

echo Launching Aptana help. Do not close this window until you are done
echo browsing the documentation.

java -classpath "$CP" org.eclipse.help.standalone.Help -command displayHelp

3a. And create a .desktop launcher file:

aptana-help.desktop

[Desktop Entry]
Version=1.0
Encoding=UTF-8
Type=Application
Name=Aptana Help
Comment=Browse Aptana Help including HTML/CSS/JS docs
Categories=Application;
Exec=/usr/bin/xterm -e ./launch-aptana-help.sh
Icon=./icon.xpm
StartupNotify=false

The script in Step 3 will only work on a Unix host, or a Windows host with Cygwin installed. The script in Step 3a will only work on a Unix host with a modern desktop environment (GNOME, KDE, XFCE, etc.)

To find the reference documentation I’ve been referring to, launch the help using the launcher script I gave you above or from Help → Help Contents in Aptana Studio, then go to Aptana Studio Help → Reference.

More downloadable documentation

The documentation packages listed below are available in PDF and HTML files for offline viewing.

Apache web server

http://www.apache.org/dyn/closer.cgi/httpd/docs/

Coldfusion application server

http://www.adobe.com/support/documentation/en/coldfusion/

• “CFML Reference” PDF
• “ColdFusion Developer’s Guide” PDF

Java

J2SE (core APIs)
http://java.sun.com/javase/downloads/index.jsp
“Java SE 6 Documentation”

J2EE (servlets, JSP, etc.)
http://java.sun.com/javaee/reference/
“Download the Java EE (number) API Specifications”

More…
http://java.sun.com/reference/index.jsp

Microsoft ASP.NET, SQLServer database, etc.

The documentation comes Microsoft’s IDEs. Be sure to install at least the Express Edition of the IDE for the particular technology you’re using, and be sure to choose to install the documentation when you get to that option.

MySQL database

http://dev.mysql.com/doc/#refman

Oracle database

http://www.oracle.com/technology/documentation/database.html
Download the 389MB archive, or browse and download individual PDF files.

Python

http://docs.python.org/download.html

PHP

http://www.php.net/download-docs.php

Perl

Documentation should have come with whatever distribution of Perl you’re using — unless you have only XAMPP. If you don’t have a full standalone version of Perl installed, go install it now. If you are using a mainstream Linux distribution, be sure that you have the package perl-doc installed.

And don’t forget:

• Your favorite IDE or text editor.
• A working copy every piece of your project(s): source code, documentation, build scripts, etc.

Please contact us as soon as possible to arrange an unnessary equipment upgrade

So, we all know the Digital TV Transition came and went, and the government imposed new rules dictating that over the air broadcasts will be digital-only from now on.  Most of us know that this doesn’t necessarily dictate any change to cable TV companies’ method of transmitting any channels from the central office to you via cable, regardless of how they receive this particular channel in the central ofice before relaying to the customer — but cable companies used the hoopla over the transition as a (well-deserved) excuse to get rid of their analog-only receivers and pack more digital channels into the same space. Here is what Comcast is doing in Boston:

• Comcast channel numbers 2 through 23 will continue to be transmitted over the cable in analog, backwards-compatible NTSC format as they always have been.
• Channels 24 through 96 which used to be transmitted in analog format are now being transmitted in digital format only.
• Channels 100 and up will continue to be transmitted in digital format only, as they have been for several years already.

If you want to receive channels 24 through 96 and you don’t have a digital Comcast TV receiver, you must upgrade now.

So far, so good, right? I am currently paying for and receiving only channels 2 through 23, which is their cheapest (unadvertised) level of service, and I will continue to receive these channels in my analog TV without any additional equipment.

Last week, I got a letter that layed out exactly what channels were being converted to digital format, and strongly requesting that I call them immediately to arrange an equipment upgrade, despite the fact that it is not necessary for me, specifically, to upgrade my equipment!

How many old folks are using the same TV package as I am and got this letter, called them, and got slammed into upgrading to a more expensive level of service?

In summary, I got a less-than-clear letter about an major service change that doesn’t affect me. Next, I have an even more irritating story.

We’re adding helpful features to our DNS infrastructure to assist you in failing to connect to your VPN

This was discussed on Slashdot, so I knew it was coming, but I do not recall getting any notice directly from Comcast about it.

In a nutshell: breaking away from standard DNS protocol, Comcast modified their DNS servers so that instead of correctly responding with “no information avaialable” to requests to map unknown hostnames to IP addresses, the DNS servers respond with the IP address of a Comcast search/ad web site

You have to be a networking expert to to notice the problem with this scheme, but to such people, the adverse effect is painfully obvious: This change will likely break the VPN client you use to connect to your company’s private network! Most VPN clients depend on the “no information available” answer; when you look up private-server-name.mycompany.com, and after your ISP’s public DNS servers report no answer, only then does the VPN client try again by asking your company’s private DNS server for the IP address.

Comcast’s DNS change was implemented to “get [you] where [you] want to go online even faster and easier than before.” Quite contrarily, I was not able to connect to my company VPN last night until I remembered about this change and fixed the problem on my end. (The fix was to temporarily switch my computer to use OpenDNS instead of Comcast for DNS, while I wait for my automated opt-out request to be fulfilled at Comcast.) If I didn’t know what I was doing, I would have had to wait at least 9 hours for my company help desk to open and help me.

As I said above, you can opt out this standards-breaking idiocy, but I wasn’t properly informed. How many hours will be wasted at different company help desks this week first, figuring out why so many clients can’t connect to services inside VPNs, and then helping them fix it?

Okular is a great document viewer, but unless you’re a Linux geek, you’ve probably never heard of it. If you are a Linux user, you probably didn’t know it was available for Window now.

For the purpose of this post, I’m considering the task of reading small PDF documents and large book-sized PDFs on the desktop. I generally don’t care about fill-out forms, DRM, encryption, and stuff like that, so I won’t evaluate such features wherever they may be available.

Here’s a brief run-down of the free PDF viewers I know of for Windows:

Adobe Reader

This is the one everyone should know about. Adobe created the PDF format in 1993; fortunately for us, some time after that, they declared it an open format that anyone could produce software for, royalty-free. They also wisely chose to distribute the viewer, Adobe Reader, for free.

Pro: Everybody else is doing it! Also renders individual pages very quickly, even when rapidly scrolling forward and backward in the document; supports annotations.

Con: Takes too long to start; no “Bookmark this page” command; can’t search within Table of Contents headings.

GSView

GSView is the reference implementation of a viewer to go with the GhostScript system from the Unix world. The Windows package for this application seems to be an afterthought.

After I installed GSView, the initial window spilled across my two displays. Then it said I had to install GhostScript and sent me to a web page were I couldn’t figure out what file I should download.

Deal breaker: I couldn’t figure out how to get it running without going off on a troubleshooting tangent.

Foxit Reader

Foxit Reader is a fairly well-known small and fast-starting alternative to Adobe Reader.

Pro: Starts up in a flash; renders pages as quickly as Adobe Reader, as far as I can tell; support annotations.

Con: It’s nagware, and the other contenders aren’t; no “Bookmark this page” command; can’t search within Table of Contents headings.

PDF XChange

I found PDF XChange listed in Wikipedia and tried it out for the first time while writing this post. Eew eew! Toolbar-itis! Next, please.

PDF XChange: Toolbar-itis

Pro: Didn’t test far enough to find any positive things.

Con: Product’s web page is full of ALL CAPS — that turns me off; nagware; toolbar-itis!

Okular

Okular is the document viewer component of the KDE Desktop Environment for Unix systems. KDE has been available as an add-on for Windows since 2008, which is quite handy if you’re a hacker living in the Unix world and you’re already a fan of any particular KDE applications.

Pro: Starts up really fast; fully searchable Table of Contents; ability to bookmark an individual page within a file; support for annotations; support many different document and image formats (PDF, CHM, DVI, GIF, JPeg, OpenOffice, etc.)

Con: To install it in Windows, you need to use the kdewin-installer and either choose the default install (most of KDE) or wade through for the one app you want; relatively slow rendering when rapidly scrolling through a document.

Sumatra PDF

Sumatra PDF is another free PDF viewer that’s been floating around for a while.

Pro: Starts up really fast; sorta searchable Table of Contents.

Con: No “Bookmark this page” command.

Dealbreaker: Can’t copy document content to clipboard.

Why Okular is my favorite

I’ve used the KDE Desktop Environment on and off in my various Linux computers, in the past few years. In that time I’ve become familiar with Okular, since it’s KDE’s default viewer for ready-to-publish document formats such as PDF.

One of the must-have features I find in Okular is strangely absent from all the other software I list above: type to search in the Table of Contents pane.

Okular: Typing in the Table of Contents pane to search only headings

Sumatra PDF lets you type on the Table of Contents pane, but after I find the first match, I can’t figure out how to search again from that point. Adobe Reader, Foxit, and others don’t even come close — as far as I can tell, you can’t search the Table of Contents links; you can only search within the body of the document, which is rather annoying when you have to wade through pages and pages of hits when you know the topic you want is listed clearly in the Table of Contents. You’re forced to scroll to the page where the Table of Contents appears as regular content, and then hope those items are hyperlinks.

Consider one of the tasks I perform most often in Okular, which is looking up built-in functions and tags in the ColdFusion reference manual. Most of the time I know the name of the function I want, so I just click on the Table of Contents pane, type in that name, and the list below is automatically filtered to show matches. Using Okular’s competitor applications, I can’t find any easy way to this.

Another important feature of Okular is the Add Bookmark command. Consider the number of “ebooks” distributed as PDF files, along with all the technical manuals, project reports, and other electronic material you might spend a long time reading; how could any major PDF viewer not have a command to bookmark a specific page?! Okular has it right in the main menu where you can’t miss it.

Okular: Add Bookmark command

Granted, most PDF viewers support hyperlinking to a specific page in a PDF file. For example you’d link from HTML by writing this:

<a href="book.pdf#23">Page 23 of 'Book'</a>

But where’s the UI command to save your spot when you come to a stopping point in an enthralling page-turner like the ODF Specification (or perhaps some lighter fare like Down and Out in the Magic Kingdom)? Adobe Reader and Foxit Reader don’t have this feature as a UI command, and I’m completely mystified by its absense after so many years.

Okular has these two features while the competition does not. In other respects, Okular is good if not spectacular. Therefore, Okular is hands-down my only desktop-based PDF viewer.

What’s your favorite PDF viewer, and why?

Windows 7 is Microsoft’s successor to Windows Vista, but since so few people adopted Vista, most people who get Windows 7 will be moving straight there from Windows XP. This includes the vast majority of corporate Windows desktops, as most enterprise desktop teams were not convinced to upgrade to Vista in the nearly three years since its release.

Personally, I am not coming from Windows XP. I abandoned my company’s standard Windows XP desktop image three years ago; I couldn’t stand all the issues of managing a Windows XP computer coupled with idiosyncratic internal corporate packaging of external packages. (If my company’s desktop team is reading: I don’t mean any offense; you know I’m just really picky.) More important though, at the time, was my strong preference for Ubuntu Linux as a desktop platform.

My dream Linux desktop at work proved to have a finite life, though. With Windows infrastructure all around me, it became increasingly tedious to fire up my Windows virtual machine inside my Ubuntu desktop for so many day-to-day tasks, particularly using my team’s preferred IDE, Adobe FlexBuilder, which still lacks a Linux release, and won’t run under Wine. So, I decided to go back to Windows, but I didn’t want to take a step back to Windows XP.

Fortunately, for the last few months, Windows 7 RC (Release Candidate) has been available as a free download. You can register and then install it for free as long as you don’t mind acquiring a real license and rebuilding your computer when it expires in March 2010. When I found out about that, I was sold. I have very little experience with Vista, but compared to Windows XP, here are some benefits I see for hackers using Windows 7:

• They managed to squeeze in some decent privilege separation/escalation. Desktop software in the Windows ecosystem has traditionally assumed the current user has the run of the system and can do whatever he wishes. This has the unfortunate side effect that if you try to setup an environment where you normally log in as an unprivileged user, you will find yourself living a nightmare with random unexplained application crashes and having to log out and log back in as Administrator when you want to install a new application. To make Windows work more like OS X and Linux desktops, the Windows team came up with a compromise, where poorly-designed applications think they are running with unlimited access, but Windows prompts you before it allows an application to make a system-wide change.
• The Start Menu and Control Panel are finally searchable, just like OS X has been for years, and KDE last year added in 4.0. (Yes, this was in Vista, but I never tried it.)
• 

  Windows 7's new Open dialog box

  The file manager, Windows Explorer, does a pretty good job of hiding those annoying kludgy drive letters from you, even in the system Open/Save/Choose Folder dialog boxes. As in OS X’s Finder, and various Linux-based file managers, you can highlight your favorite folders in the main navigation pane, so that most of the time you don’t have to go to the My Computer folder and deal with drive letters.

• Internet Explorer 8. Wow. This is what IE 6 could have become if anyone at Microsoft had been allowed to do serious development work on it. Tabs, streamlined window layout, and a developer tools screen that I find easier to use than Firefox’s Firebug.
• A killer new object-oriented command shell environment where input and output streams are hierarchical objects and collections instead of plain text streams. Not specifically a Windows 7 feature, but I’m putting it in here because I’m excited to try it out when I get a chance.
• When you connect to the “Console” running the main desktop, remotely, Windows first saves all your window positions, then rearranges them to fit in your terminal’s geometry. When you disconnect, saved positions are restored for any processes that are still running. Yay! — actually it seems like there are little wins like this all over Windows 7

While Windows 7 will undoubtedly be a crowd-pleasing success, I still have a lot of issues with being a hacker and developer living on a Windows box, for instance:

• You can’t mount SFTP servers locally and have other ad hoc pluggable filesystems à la FUSE. Actually you can. As I was writing this blog entry, I figured out how to get Dokan and Dokan SSHFS working in Windows 7. Yay! I need to write up a howto (but in the mean time, see this bug.)
• The window manager still kinda sucks. You can’t alt-click-and-drag to move Windows like in GNOME and KDE; that’s a third-party add-on. I can’t find a way to get window edges to snap to other window edges.
• No built-in clipboard history.
• No built-in SSH server or client for remote login and file sharing. SMB (Windows file sharing) is irritating, and other operating systems don’t play well with it.

And there are some issues aren’t particular to Windows 7; they are really problems with the Windows ecosystem and not something a new Windows release could resolve on its own:

• Standard package repositories — Yes, Windows has had a respectable package manager for a long time, but what about a centralized repository? I want to open up my package manager, type in the name of an application or a keyword, and then install the application with one click. I want automatic updates to come from the same place, instead of numerous annoying third-party Systray apps that phone home every week looking for updates in their own special way. I don’t think Microsoft is interested in dealing with the political issues of managing and editing a catalog like this, and third party attempts suffer from a chicken-and-egg problem: why should one developer buy into your package management system when no one else has yet?
• Where do you find Windows software? Seriously. After spending most of my time in various Unices for the last several years, I can’t for the life of me figure out how to find Windows applications to solve any problem I might have. If it’s a good open source app available for multiple platforms, it’s probably listed in Freshmeat, but if it’s not free or not multi-platform, I don’t know where to find it. How do you sort through Google spam and find respectable applications you don’t already know about? How can you trust what you find?

So here’s how I customized my Windows 7 desktop to work my way (all of these products are free-as-in-beer or better):

• Desktop environment add-ons: ClipX for clipboard history; Taekwindow for alt-click-and-drag to move windows; WinSplit Revolutioon for better window management (still no snap-to-edge); entire Windows port of KDE applications, which includes my favorite PDF viewer, Okular.
• Desktop accessories: CD Burner XP Pro and Microsoft Virtual CloneDrive for compiling, burning, and mounting CD/DVD images; 1by1, a minimalist, database-free music player; IrfanView, a simple no-frills image editor/converter.
• Cygwin for common Unix utilities like Midnight Commander, GPG, and bash.
• Remote access: Portable RDP client from TSplus because the Microsoft client was thrashing my CPU (go figure!); PuTTY for remote text login; FileZilla for file transfer; NoMachine NX Client for remote desktop login to my file server at home.
• Zim notepad/outliner (Windows version is available here).
• Virtualization: VMWare Server 2, after much hemming and hawing (I should write this up as a separate blog post), running Ubuntu 9.04. Seamless remote desktop login to the virtual machine by using Xming and XFCE Desktop. See my howto.
• TrueCrypt for transparent encrypted filesystems.

It was a lot of work, but I’m pretty happy with the results, and I’m definitely not a Microsoft-hater anymore.

In future posts, I will highlight some of the things above (for example VMWare 2 or Okular) that I find particular interesting. I also plan to review books and web sites. I also hope to be writing a lot about information portability and security, from the perspective of your personal data.