XMoby's Geek Corner

Having fun with application indicators in Unity

2011-04-30T15:05:00.000-04:00

If you, like a whole lot of other people (if the download speeds on the Ubuntu servers are an indicator), just installed the latest version of Ubuntu, you probably realized that Unity has a... different ways of doing things. Here, I'll just focus on the application indicators (a.k.a. the system tray).

First thing you'll realize, just after the fact that the indicator is now located in the top-right corner of the screen, is that a lot of (legacy?) applications do not show their icons in the tray. This is "normal": Unity offers a new API for indicators, and (by default) requires the applications to be coded specifically for Unity. What that means is, most applications that are not actively supported most likely didn't switch to that new API, and as a consequence, are not present in the tray. Also, some applications where not using the tray, but were GNOME applets themselves. Those, again, have been booted out the Unity party. What a bummer, especially if you had a kick-ass panel setup.

What to do then?

The current state of my tray

White-list indicators

One thing you can do is to white-list existing indicators. Eventually, the apps you use may be ported for Unity compatibility, but why wait? In the meantime, just indicate to Unity that some (or even all) applications should be allowed to show in the system tray.

To do so, you'll first need to install dconf-editor from the dconf-tools package. You can get it from the Software Center in Ubuntu.

Once it's installed, launch it (either by pressing Alt+F2 then typing dconf-editor, or through a Terminal). In the application, navigate to Desktop -> Unity -> Panel, and modify the systray-whitelist value. You can add each application individually (personally, I added 'Last.fm' and 'Dropbox'), or replace the whole thing by ['all']. Be aware, however, that some applications might mess a bit with the tray, so if you happen to experience some weird behavior, be sure to revert back to the old value. There's even a nice "Set to Default" button in dconf-editor, so that you don't have to remember what that default was.

Using dconf-editor

Oh, and make sure to at least log-out then log-in for the change to take effect.

Install new indicators

Sometimes, an indicator you were using is not compatible for Unity, but a nice alternative is available. Or maybe you just want to try something new. There's indeed a lot of nice indicators to use. Here are some of the nicest ones, in my opinion:

Caffeine: Allows to enable or disable screensaver and sleep mode, either manually or depending on some applications (e.g. if Banshee is open, don't sleep).
Ejecter: A simple icon that allows to easily unmount external drives, such as USB sticks, CDs, external hard drives. Can sometimes by faster than doing it in Nautilus.
Indicator-Weather: A nice, configurable weather indicator, with multiple locations and sensible preferences.

References

Get a Windows 7-like taskbar in GNOME with DockbarX

2011-01-19T23:26:00.000-05:00

One of the first thing I *really* liked when I first tried Windows 7, was the taskbar. This launcher-switcher hybrid that allowed me to simply click on the unique Firefox (or whichever) application icon to pop it, without having to think "ok, if it's already launched, I should click on the taskbar, but if not, I'll have to click on the launcher". I didn't really realized how stupid it was to have, on the same bar, a lot of duplicate icons (because what is set as launcher/shortcuts is often also what is running most of the time). Both icons did have the same role: bring the application to the front. It shouldn't matter whether it was already launched or not. I think the way it's done in Windows 7 is just fantastic*. Add to that: the jump lists, the recent and bookmarked documents, the previews... Well, it's just perfect IMO. If you absolutely don't know what I'm talking about (and my messed up description probably didn't help), just take a look at this.

That being said, even if the Windows 7 taskbar is the most perfectest perfect taskbar, that was not enough for me to simply ditch Ubuntu. Instead, I tried to find something that would work a bit like that taskbar. I experimented with a lot af different switcher GNOME applets, but none really did it... Until I found DockbarX. Since that day, I'm blessed with app selection goodness and intuitive UI. Life is great.

You can also get a piece of that heaven, by following the instructions here (download link, PPA, extra themes and all). Then, simply replace your standard application launcher/selector/taskbar applet by the DockbarX, and you're ready to go.

Now, I just need to find a good Snap-like thing. This just didn't do it. I may also just wait for Compiz 0.10... Please wave in the comments if you know a good solution for this.

* By the way, I don't know if Microsoft were the first to come with the idea, or if it came for Apple, or from somewhere else. And quite frankly, I don't care.

Dropquest 2011: Get a free 1GB from Dropbox (bring your thinking cap!)

2011-01-19T22:47:00.000-05:00

You probably already know about Dropbox. If not, let's say it's self-desbcribed as a "software that syncs your files online and across your computers". I'll also add that it can sync with mobiles (iPhone, Android and Blackberries at least), and allows for some neat tricks (I'll share some with you... eventually).

So what I'm talking about here, is getting a free GB of storage, simply for following some steps. Mind you, I didn't say simple steps. In fact, it's a full-on scavenger hunt, with all the head-stratching it implies. A game that spawns the whole interwebs, and even more. A game you probably won't finish in ten minutes (except if you cheat and use some FAQs, but seriously? You're certainly better than that) At least this time, you won't have to bother your friends and family to subscribe for a 250MB referral bonus per user you bring on-board.

What are you waiting for? Get that cloud-located gigabyte right now, here: https://www.dropbox.com/dropquest2011

P.S. As a totally shameless plug, if you don't already have a Dropbox account and want to create one, please use this referral link: http://db.tt/OGFrNxO. Both of us (that is, you and me) will get a 250MB bonus, in addition of the basic free 2 GB. Add to this the 1 GB mentionned in this article (you've been reading, right?), and you'll start with a nice 3.25 GB. Ain't that nice?

Enjoy a Mont Blanc-like writing feeling, for a (very small) fraction of the price

2010-04-29T21:30:00.001-04:00

I stumbled upon an Instructables article a couple of days ago, and since then, I was obsessed to give it a try. Even though I'm a technology geek, I just can't deny my obsession with pen and paper.

The whole thing consists of taking a simple Pilot G2 Pro pen body, and replacing the ink shaft by a nice Mont Blanc one. The refills are surprisingly similar, just requiring a lit trim at the top of the Mont Blanc one. The details are in this article, with a little bit of humor threw in.

And for the records, this also works with a standard G2 (non-pro) pen, and probably any other that would normally take a G2 refill. Just saying, because I know I wasn't able to find any G2 Pro in local stores around me, but standard G2 were aplenty.

Enjoy your while music collection from (almost) everywhere... even your iPhone or Android device!

2010-04-29T13:00:00.001-04:00

Over the years, you may have grown quite a huge music (digital) collection. You know, the kind of collection that doesn't quite fit into your portable device storage space. Smartphones and even laptop hard drives have their limit, or you may prefer to keep an empty buffer for other things.

Well, if you got an always-on computer at home that is storing your whole collection (or if you're ready to do a move in that direction), you'll be able to free your music from the confine of your walls, thanks to Subsonic.

The setup requires that you first install the Subsonic server on your always-on machine. The software is free, and available for Windows, Mac and Linux. For the download and details on installation and configuration, just head to their site.

Once the server is setup and running good locally (you'll be able to test it in your web browser), it's time to make it accessible from the Internet. Don't worry, your precious music (and bandwidth) will be protected from strangers by the built-in Subsonic user accounts system.

First, you will probably want an easy-to-remember URL. If you opt to donate to the Subsonic project, they'll happily provide you with a user.subsonic.org address. Or, you may want to take a look at my strangely convenient previous article about a dynamic DNS.

Next, if your home network is running behind a router, you need to make sure the Subsonic traffic gets correctly routed. Just forward the port 4040 to your local Subsonic server, or any other port if you changed the default configuration, which might have been necessary if your ISP blocks some ports or something like that.

At this point, you'll be able to access your collection from any browser, everywhere the connection speed is at least decent. But what about when it's a bit less speedy? In that case, you can create a specific Subsonic user account on your server, on which you'll set a speed limit. By activating and configuring this option, Subsonic will transcode (i.e. resample as it streams) any file to the specified bandwidth. This will allow you to better control you own upload bandwidth, and also limit the download bandwidth you be using on the receiving end.

There are a couple more options you may want to take a look at on the server. For example, you can associate a Subsonic user account to a last.fm one, so that all songs played are scrobbled. There is the possibility to let each user select it's own browser skin. You can also ask Subsonic to retrieve all your missing albums cover art. There's really a lot to explore here, so enjoy!

Ok, now that everything is configured on the server side, it's time to look a the options we have on the client side. Sure, we can stream from the browser, but it can be a bit clunky, especially on a smartphone's small screen. If you're running on an iPhone or Android-powered device, you have access to native player applications that will connect to your Subsonic server, and deliver you all that audible goodness. A simple search on "subsonic" in the device's integrated app store will show you what's available. At the time of this writing, two $5 applications are available on the iPhone, and a free one will satisfy those Android users. There's also a desktop application, for those interested.

All is not perfect, however. To make sure you don't abuse your phone data plan, or simply to allow your 3G connection to keep up the pace, you'll probable need to down-sample what you're listening to. Also, at least on the iPhone, you won't be able to stream music while you're doing something else, thanks to its no-multitask "feature". And there's always the risk of a connection problem, preventing you from accessing your collection. That being said, just make sure you still keep a couple of classics on your device for those cases, and you'll have the best of both worlds.

No need to remember your public IP address

2010-04-26T13:00:00.001-04:00

There are a lot of reasons why you may want to know you public IP address. Maybe you set up an FTP server, so you can always access your files when you're away. Maybe you collaborate on a project, and want to host a source control server. Maybe you installed some other service, that you would access through a web interface. Or, maybe, just maybe you refrained doing any of this, because you couldn't bother to take note of your public IP address each morning before leaving home, just in case your service provider changed it overnight.

Well, there's really no need to remember your public IP address (sounds like the title of this article, doesn't it?). For quite some time now, it's possible to freely register a subdomain (for example, myname.dyndns.org), and link it to your own setup.

How does that work? It's quite simple, really. There's no redirection, port forwarding or anything like that. The subdomain you register will simply resolve to your public IP address. Consider that the subdomain will act as an exact synonym for your public IP address. That's it, nothing more, nothing less. This will not create a web server or anything though. This will only allow you to easily access any service you provide (and have set up) on your home network.

Once you register your subdomain, you'll need to keep it up-to-date with your public IP address. Depending on the service you choose, you may have to run an application in background, that will detect any change, and inform the server of it as it happens. If you're lucky enough, you may even have a built-in option in your router configuration that does exactly this, so there's no need to lose some precious CPU cycles and kilobytes of RAM on one of your computers. I know some Linksys routers (WRT54G and WRT310N, at least) offer the option for dyndns subdomains. Whichever method you use to keep the service updated, once it's set up, you can forget about it, and let it do it's job.

There are a couple of offerings for this kind of service. However, there are really no options or anything, so it's really a matter of reliability, and the availability of update tools, that will direct your choice. I personally opted for dyndns.com a couple of years ago (the free service), and never looked back.

Holy box of rhythm, Beatman, I can now sync my iPhone on Ubuntu!

2010-03-13T19:20:00.000-05:00

Yup, you read right: it's now possible to sync the music on your iPhone directly in Ubuntu, with Rhythmbox. Personally, I don't have a need for video transfer, and all my other stuff (contacts, mails, tasks list, etc.) is already synchronized automatically through the Internet, so that's a really good news for me.

I may be a couple weeks late, but I just read the article over at OMG! Ubuntu! (by the way, I added a shortcut to that site in the "Links of Interest" box, bottom-right corner of this page). If you want a (very) more detailed how-to, with screenshots and all that, you can head to the original article, at MakeTechEasier. I won't repeat the whole how-to here: just use one of the two links in this paragraph to enable this goodie on your own machine. It may seem complicated, but seriously, it's a matter of minute(s).

I just tried transferring a 13-track, about 55-MB album, and it took about 5-10 minutes. Not thaaaaat fast, but then again, it beats the über-long synching that usually comes with the whole process in iTunes. Plus the fact that, as a Ubuntu user, I normally had to launch VirtualBox, my Windows XP virtual machine, then iTunes, then start synchronizing stuff... Saving me a lot of time.

To send music on your iPhone, simply drag and drop what you want, in Rhythmbox, from your library to your device (on the left-side bar). A little progress bar will appear on the bottom, so you know when it's done... almost! I noticed my iPhone going into "Synchronizing" and "home screen" mode, back and forth, a couple of times. I don't know why it did that, but I wait for the back-and-forth to stop, then I went into the iPhone's iPod app to validate all was sent. At that moment, it was OK to eject the device (right-click the device name, then select the appropriately named option "Eject"), then disconnect. Joy!

Just a quick note, as seen in the comments of one of the how-to's: make sure to disable your iPhone's password, or else, it won't be detected in Rhythmbox.

Avoid procrastination by using a work-only firefox profile

2010-02-28T14:40:00.002-05:00

Most of us have been there before: you begin to do some work, which at some point involves browsing the web for some information or another. While you're there, your time-wasting bookmarks poke at you, an email or RSS feed notification pops, and before you know it, you lost twenty minutes, an hour or even more. And, mainly, all motivation to finish what you started is gone.

In an open-space work environment, where people can see at any moment at you're doing, this is less of a problem; there's some kind of self-regulation taking place, and you can hardly abuse. However, if you got an isolated workstation, are self-employed, or working on side projects, there's no one to keep you on track. If you regularly find yourself in such a time-wasting situation, like I am - this blog is a living (dying...?) proof of it, as are a lot of other personal projects -, you'll want a solution that will force at least a bit of discipline into your work time.

There's a lot of Firefox plugins that can help you stay on track, by blocking access to certain sites depending on the time, etc. (see LeechBlock), but I found that was not enough. First, it assumes a regular work schedule, which is not the case with my side projects. My shortcuts still remain there, prompting some kind of reflex to open them (even though they are blocked). My email and RSS notifications sometimes pop. I can glance at the weather widget a bit too often (I know it's snowing, damnit!). So, there's too much I can do that's not work.

Following the advice in found in a LifeHacker article, I created a "work-only" profile in Firefox. It may seem an overkill, but believe me, it's easy to setup, and does a really nice job.

What is a Firefox profile?
You can consider a Firefox profile as if it was a completely different Firefox installation than the one you're normally using. That means it has it's own set of bookmarks, extensions, navigation history, password, etc. Anything you do in one profile, never affects the other one(s).

Nice. Now, how do I create and access that profile?
First, you need to access the Firefox Profile Manager. On the command line, this can be done with the following command:

firefox -ProfileManager -no-remote

The -no-remote option is necessary only if you have an instance of Firefox currently running (which is most probably the case if you are reading this article).

From there, you can create your new profile.

Now, to start Firefox using your newly created profile, simply launch it using the following command:

firefox -P NAME_OF_THE_PROFILE

Note that is there's already an instance of Firefox running, the command above will use the profile of that instance, and not the one you specified. To override this restriction, simply add the -no-remote option at the end of the command.

To facilitate the use of that new profile, without using the command line, just create a launcher with the whole command.

OK, that profile is completely bare. What should I put in it?
Personally, I go with a couple of productivity-enhancing, twaeking extensions, like Tab Mix Plus and FireGestures. Also, Adblock Plus does a nice job reducing distractions (annoyances) on a lot of sites. Also, think of any job-related extensions you might want, like Web Developer, Firebug and the like.

And, last but not least, LeechBlock. This can block access to all the sites you specify. You know where you normally lose track of time (digg anyone?). Make sure to configure the extension as tight as possible. Remember, this profile is to be used when you plan on working.

Also, to make your job easier, set all the bookmarks, search engines, and remembered password that can help you work.

That seems good. Anything I should avoid?
Sure! First, make sure you don't have any bookmark or remembered password for sites you should not use. But then again, you blocked them, right?

Also, avoid any pop-up extension, like mail and RSS feed checkers. Even when you are the most focussed, it doesn't take more than a little "You got new mail!" sliding window, or a huge count of unread items, to make you lose your precious concentration.

This is just plain evil...

And what's next?
Hopefully, a more productive environment. On my side, let's see if that leads to a bit more articles on this blog.

Also, if you are a bit of a jack-of-all-trades person, working on a couple unrelated projects (e.g. web development, writing, etc.), you could create more than only one "work" profile, and tailor each profile to its tasks. If you're willing to spend a couple of minutes to tailor your work environment(s), you can probably save a lot more in the long run.

Just rescued my (software) RAID5.

2010-01-02T14:00:00.001-05:00

Yesterday, after rebooting my home server, I noticed I couldn't access any of its data, which sits on a normally fine software RAID5 device. When trying to mount the device, I always got the following error: "mount: /dev/md0 already mounted or /mnt/share busy".

I tried rebooting: fail.

I messed a bit with my /etc/fstab config file: fail.

Suspecting the mdadm config file could be the culprit, I uninstalled, then reinstalled mdadm, to make sure it regenerates its config file: fail again.

After some long (and panicking) minutes, I found this savior's post:
http://matthew-garman.blogspot.com/2009/12/device-already-mounted-or-mountpoint.html.

Just a quick note though: make sure to remove ("comment out") the necessary entry in your /etc/fstab config file, then reboot, before attempting to run e2fsck. If not, the device may be "busy", probably trying to be mounted, and the scan won't launch. Then, after all is repaired, restore your original fstab file content, and remount your device. Voilà!

Easily manage multiple launchers in a GNOME panel with quick-lounge-applet

2009-11-06T21:06:00.005-05:00

If you like to have a great number of application launchers on your GNOME panel, but don't like the way you have to move each and every one of them when you re-organize your applets, quick-lounge-applet is for you.

What is it?
quick-lounge-applet is a GNOME panel applet that can manage a list of launchers, with ordering, icon and name selection, separators for sections, etc.

Why use it?
Since you'll be using only one applet for are launchers, instead of an applet for each single launcher, you'll be able to move this as a whole, should you want to re-organize your panels. For example, if you have 15 launchers on the top-left panel, and would like to move them to the bottom panel, it would be cumbersome with actuals launchers. With quick-lounge, it's a one-step operation.

Also, you'll be sure that the launchers get all evenly spaced, and you won't have any trouble adding a new launcher between two already existing ones.

How to use it?
In Ubuntu (and possibly other distros), you can install the quick-lounge-applet package.

sudo apt-get install quick-lounge-applet

Once done, right-click on the GNOME panel, where you want to install the applet, and select "Add to Panel...". Select "Launchers List" from the list, then click "Add".

To add, remove and edit existing launchers, right-click on the applet you just added, where there is no launcher icon, and select "Preferences" from the popup menu. From there, I'm sure you can figure out what you have to do.

That being said, do you think such application launchers are still relevant? Personally, I don't see myself getting rid of them soon, but who knows?

Always keep a terminal at hand

2009-08-03T23:45:00.001-04:00

A lot can be done in a Linux terminal. You can launch any "pathed" application or scripts, and thus get info, activate features and whatnot. It then makes sense to always have a terminal window ready for action, right?

In this article, I'll explain the two different methods I'm using, depending if I want to do light-weight or heavier stuff.

Global shortcut
Since you'll be using both hands on your keyboard when working in the terminal, it does make sense to start the work with a first keystroke. To know how to enable a global hotkey shortcut, just read this other article of mine. In GNOME, the command to launch would be gnome-terminal.

I use a dedicated terminal when I have a lot of terminal work to do, like testing commands for an work-in-progress script, or when moving or ch'moding files, or any other more-than-one-or-two-commands task.

Also, to avoid having to deal with a ton of open windows, don't forget you can open multiple console tabs in the same window by using the Ctrl+Shift+T shortcut while in gnome-terminal.

tilda
tilda is a GNOME quake-like console. You can "pull it down" by pressing a dedicated shortcut (I like to use F12), and hide it with the same hotkey. And, just like in gnome-terminal, you can work with multiple tabs, again by using the Ctrl+Shift+T shortcut.

You can install tilda in Ubuntu from the universe repository (which may not be enabled on a default installation). From a terminal:

sudo apt-get install tilda

You can configure a couple of settings either by right-clicking the tilda console and selecting "Preferences", or by launching tilda in configuration mode (-C command-line switch). You'll be able to select the global shortcut that will show or hide tilda, the position and size of the console, if you want tilda to start hidden (nice if you set it to launch with your GNOME session), etc. Also, if tilda ever segfaults on you when you try to launch it, start it in configuration mode. It always worked for me.

For the ones of you who prefer KDE over GNOME, you'll probably prefer using an alternative more suited to your environment, namely, yakuake.

I prefer this unobstrusive, not-many-lines-displayed option (tilda) when I want to do a quick operation, like mount or unmout my file server shares, install an application through apt-get, or something like that. The tilda window shows fast, gets the job done, then gets out of the way.

Other options
There are may more ways to have a console ready at hand. You can have a terminal as your desktop background, add a gnome-terminal launcher to any of your panel, auto-launch a gnome-terminal with your session, etc.

Personally, I use the two methods I described, and they mostly fill my eager console needs. What about you? How do you keep a terminal at hand? Don't be afraid, share your wisdom and tricks, we're all listening.

Rapidly start a common task in Gnome with system-wide hotkeys

2009-07-07T12:30:00.000-04:00

Not all of us are heavy mouse users. Even though "Start menus" and quick launch icons are quite useful, it's not always efficient to have to leave the keyboard to reach the mouse just to launch an often-used application. The same way you probably use the Ctrl+S shortcut more liberally than the Save icon (or even worse, the File > Save option), a simple two- or three-key combination can save you precious milliseconds for those repetitive tasks.

In GNOME (the default desktop environment for Ubuntu), there's a way to create up to twelve global keyboard shortcuts. In fact, there are not only A way, but MANY ways. I'll show you some in this article.

How to edit global shortcuts
In the Gnome Configuration Editor
You can start the Configuration Editor from the Applications > System Tools menu (you may need to activate this item in the menu editor), or from a console (gconf-editor). If you never used this tool, you'll find it's quite similar to Window's regedit: you have a tree on the left, and values related to the item selected in the tree, on the right.

To edit a global shortcut, you'll have to change two values: the key combination, and the command. Both values are linked by the same command number (noted as (N) just below).

/apps/metacity/global_keybindings/run_command_(N): The key combination to trigger the command. You can refer to other values already present in the Configuration Editor page. For example: w. (Note: represents the "Windows" key)
/apps/metacity/keybinding_commands/command_(N): The command to launch when the key combination is triggered. For example: firefox.

Any change done here is instantaneous. If a shortcut does not seem to do what it's supposed to, enter something like gcalctool as the command_(N) value, to see if at least the key combination gets registered.

In the CompizConfig Settings Manager
If you are running Compiz, and have the CompizConfig Settings Manager (CCSM) installed (you should, really), there's an easier way to manage your global keybindings. First, make sure to install compizconfig-settings-manager, either from Synaptic or the command line:

sudo apt-get install compizconfig-settings-manager

Now, start the tool by going to System > Preferences > CompizConfig Settings Manager. In the General section, open the Commands item. There, you'll see a first page,"Commands", and a second page, "Key Bindings". Just like with the Gnome Configuration Editor, enter a value in two linked fields to activate a global shortcut. As simple as this!

In Ubuntu Tweak
Even simpler the the previous methods, this one also involves installing a new application: Ubuntu Tweak. From the repos, install ubuntu-tweak, using either Synaptic or the command line:

sudo apt-get install ubuntu-tweak

Once installed, launch Applications > System Tools > Ubuntu Tweak. In the Personnal > Shortcuts section, you'll find the most intuitive way to edit your global keybindings yet.

In a terminal, through a shell script
For the console-lovers out there (I'm one of you guys), I created a nice shell script to manipulate the GNOME's global keybindings. Like all my (future) scripts, it's available on the XMoby's Geek Repo (direct link). Install in your ~/bin folder, and enjoy!

My personal shortcuts are:

<Super>q to launch a terminal window (gnome-terminal)

<Super>w to launch my Web browser (firefox)

<Super>x to launch my file eXplorer (nautilus)

By the way, <Super> (or sometimes <mod4>) is the "Windows" key on most keyboards.

Did you set up your environment with global hotkeys? If so, what launches what? If not, why? Please share your own tippidy-tap launch tips in the comments section below.

Find an (possibly better) alternative to (almost) any existing app

2009-07-03T00:00:00.001-04:00

I just tumbled on this site, and I thought I might share it with you. AlternativeTo allows you to search for an application you know, and find alternatives that might better suit you.

For example, suppose you like Photoshop on Windows, but need to find an alternative on Linux. Just search for Adobe Photoshop, and you'll be brought to a page describing what is Adobe Photoshop (as if you didn't know), and a list of alternatives people proposed, sorted by number of "I Like" votes. Each entry in the list specifies for which platform the alternative is available (Windows, Mac, Linux and Online/Web-based), and provides a link to a page similar to the one you're on, but for the alternative. That means you can navigate from one alternative to another to another (and on and on), until you find something of interest.

Personally, I plan to use this site to find alternative to applications I love, but still find imperfect. When there's only one or two show-stoppers on an otherwise almost-perfect tool, this site may prove quite useful.

Also, mobile phone users are not left empty-handed, as a specific "Mobile" section is also available. That way, no need to navigate endlessly in the iTunes stores and read tons of descriptions (that tries more to sell than actually describe the product most of the times) to find what iPhone application might suit you better than one you use or heard about. As with the previously-described Desktop section, each application entry tells you on which platform it is available (so you don't try to find an Android application in the iTunes store, for example).

Did you ever had the need to find a better tool than the one you were using? Did an OS or mobile phone change ever forced you to find an alternative to a loved application you could not use anymore? Feel free to share your experience in app-switching in the comments section below.

Shell Scripting Basics

2009-07-02T22:00:00.000-04:00

Considering a lot of future posts will have something to do with shell scripting in Linux, I thought I might at least provide a little primer on the subject. It's by no means complete, but has at least the merit to be accurate.

Oh, by the way, this article will be mainly oriented towards GUI users (e.g. Nautilus, gedit, etc.). If you're more accustomed working in a console, you probably know enough about shell scripts and don't need this article.

What is a script, and why should I care?
Basically, a shell script is an easily modifiable executable file. No need to know complex programming languages, or to compile, or to install external interpreters. In fact, such a file is created by editing commands you would normally enter in the terminal, and is ultimately run as interpreted by the shell behind the terminal. It allows for some conditional and/or repetitive execution of commands. And, just like in the terminal, it's possible to call any other executable file from a shell script.

What's fantastic with shell script is their versatility. You can make it do (almost) anything! You can automatize repetitive tasks, such as backups, diagnostics, etc. You can also build whole new features, using a mash-up of existing tools. If you still don't quite get it, don't worry. I have a couple of interesting scripts coming up (in future articles, though).

How do I create and run a script?
Creating and running a script is a simple process.

First, create an empty file. In Nautilus, right-click in the folder where you want to create your file, and select "Create Document > Empty File". Choose a name. Even though it's absolutely not necessary, it's not a bad idea to name your file with a ".sh" extension (e.g. "myscript.sh"). You could put any extension, or even none at all, since the file type is detected from the content, and not only the file name.

Now, open your file in any text editor. By default in Ubuntu, it will be in gedit (aptly named "Text Editor" in your applications list). We'll do a basic script here; enter the following text in the file, then save and close it:

#/bin/sh
zenity --info --text='Hello world!'

The first line indicates which interpreter should be used. In most cases, you'll want the default shell to interpret your script. The next line displays a message box (will not work if you are running in console-only mode).

Before trying to execute the script, you'll have to allow it to be executed. This has to be done manually, for security reason. In Nautilus, right-click on the shell script file, then select "Properties" (it's at the complete bottom). In the "Permissions" tab, tick the "Allow executing file as a program" checkbox, then click the "Close" button.

You are now ready to test your script. To do so, do as you would do with any other executable file: double-click! You should see a classic "Hello world!" message box. Congratulations, you (probably) created your first shell script!

Where do I store my scripts?
Shell scripts can be stored anywhere on your hard drive (or USB stick, or wherever you can store files). However, to allow easy access from the terminal, you should store any user script in /home/your_user_name/bin/ (you may need to create the folder if it does not exist), and your needs-admin-privileges scripts in /usr/local/bin/.

Where to go from here?
That was really a very brief introduction to the fascinating world (yes, really!) of shell scripting. As I said, it may not make any sense right now, but you'll learn to appreciate it.

If you are curious, and want to experiment a bit by yourself, by all means do it! Here are a couple of sites and articles that may help you to get started:

Wikipedia's article on Shell script
Linux Shell Scripting Tutorial - A Beginner's Handbook (a personal favorite)
Steve's Bourne / Bash shell scripting tutorial
Wicked Cool Shell Scripts (a couple of interesting examples)

And, as always, a quick Google search can send you in the right direction.

Feel free to add your own links in the comments section below!

Obligatory presentation

2009-06-28T22:30:00.001-04:00

Hello all! Before I begin to fill this blog with a lot of stuff (and mainly, before you decide if it's worth something or not to you), I'll take a couple of minutes to explain what you're likely to find (and not find) here. The "About" box, in the upper-right corner of this page, is a quick and useful intro. (Go read it, I'll wait... Done? Fine, let's proceed then!)

So this blog is all about the tricks I find useful for myself, in terms of productivity and/or technology. You'll find information, advices, how-to's, tools (both home-made and more official)... This being said, all I'll publish will be solutions that work for me, but it may not suit you at all. Or maybe it will offer you a nice starting point, from which you'll be able to come with your own better-than-the-original solution. Or maybe it will be something you had running for a while, but then decided to switch to something you consider better. In any case, please don't be afraid to contribute with your own experience. That's the main point of this blog: finding the best solution.

Naturally, I'll be focusing on the tools I'm using. That's where a couple of you may want to leave the boat. Without any order of preference (those tools all work best when brought together anyway), I'm using: Ubuntu Linux on a Fujitsu laptop, and on a home server; an iPhone 3G; Google services (gmail, calendar, docs, reader, blogger...); other cloud services (DropBox, SpiderOak). When choosing a software, I always try to go with a multi-platform solution. Please note that, even though I'm a Linux user, I'm not a Windows (or Microsoft) hater, nor do I despise non-free software development. After all, I'm using some paid-for services and tools, and I'm not ashamed of it. You won't see me bashing a product just for the fun of it, and I expect the same level of respect from anyone contributing in the comments. In fact, the only reason I may moderate the comments would be a clear lack of respect for anyone/anything (well, that and legal reasons, if necessary). That doesn't mean you should not voice your opinion, on the contrary. Please, just do it in a civilized manner (I don't know you personally, but I think you're able to do this).

And a final remark, for those who didn't notice this when reading this article: English is not my primary language. I'm publishing in English, so I can reach a larger audience, and share the love to more of you people. If what you read is unclear, please tell me so. I'm trying really hard (really!), but I'm not perfect (really!).