gbtsco Bluetooth Package

Originally uploaded by sogrady

This particular HowTo comes with a number of caveats.

1. I don’t particularly like Skype; the Linux client disconnects me continually on multi-party calls, the lack of a standardized protocol, etc.
2. The setup here has a lot of moving pieces: hardware, kernel modules, libraries, packages, and so on. In other words, just because this works for me doesn’t mean it’ll work for you.
3. The setup here is also not fully tested. I have working audio into the headset, but I haven’t yet tested the microphone out. My EDGE bandwidth here isn’t good enough for decent Skype, and besides, everyone’s asleep.
4. The setup here is also still a bit clumsy. Unlike working with your phone, which simply pairs with the Jawbone once the latter is activated, this requires some activation steps before usage. You may be able to further refine the process, but at the moment it’s involved.
5. It’s possible that monkeying with this stuff could destroy your headset, your laptop, burn down your house, and rob you. Or maybe not. Either way, you’ve been warned, and I take no responsibility for what you do with this. And so on.

All that said, this does provide what I believe are basic working instructions to first pairing and then configuring your bluetooth equipped Linux desktop with your bluetooth Jawbone headset.

The Background

Historically I, like most of the Skype users I know, have used the standard pluggable headsets for calls. These have the advantage of working more or less out of the box, even on Linux, and being relatively inexpensive. The disadvantages, however, are many: they add another set of wires to your kit, they don’t have the sophisticated noise cancelling abilities that the Jawbone does (it’s not marketing, it actually works – except for wind noise), and – worst of all – if you’re already committed to carrying a bluetooth headset for your phone, they add a redundant device.

The ideal solution, then, is to carry just the simple bluetooth headset for both phone and PC usage. The trick is making that work. Here’s how I’ve done that (see the caveats above).

Pairing

The first step is to “pair” your headset with the machine. Initially, I thought this would be easy as the Ubuntu wiki’s instructions make it sound not terribly complicated, but that didn’t work for me. As nearly as I can determine, kbluetoothd – the recommended pairing mechanism on most HowTo’s I’ve seen – has been deprecated in Gutsy (see the thread here).

Fortunately, I stumbled across Stéphane Graber’s gbtsco package, which will pair the headset fairly easily. Install that via the .deb file he makes available, and just to be safe open Synaptic, do a search for bluetooth, and install everything it turns up. Not saying that’s the right approach, just that that’s what I did.

Once that’s accomplished, fire up gbtsco by typing sudo gbtsco in a terminal window.[1] That’ll pop up a window that looks like the one inset. Next, make sure your Jawbone is off – i.e. not blinking white – and hold down the black button that says “Jawbone” for three seconds. That’ll put the device into pairing mode, during which it will blink red and white alternately. Now, click “Check for New Devices” and the Jawbone should show up.[2] Select Jawbone from the list and click “Use as Active Headset.” The first time you do this, you’re likely to get a flashing permissions icon asking if you want to grant the headset access to audio. Click yes, and if prompted for a code, give it 0000.

You should be paired. To test it out, try playing a WAV file to the headset by typing aplay -B 1000000 -D plughw:Headset /path/to/a/local/wav/file [3] – it should come through loud and clear.

Configuring

With the headset paired, what remains is telling Skype about it. This will likely be different from client to client, but generally you want to open options, select “Sound Devices,” and make sure that the Sound In, Sound Out, and Ringing profiles are cut over to your headset. Note that at least for me, there are two BT Headset options: one that reads (hw:Headset,0) and one that reads (plughw:Headset,0). The latter works for me, the former does not, but your mileage may vary, as always. See my configuration here.

With those changes made, try a Test Call and see what happens. Hopefully you’ll hear it ring, connect, and a voice will speak back to you.

Questions, Problems, Issues

As I’ve just figured this out, the odds are good that I can’t help you if this doesn’t work. I’d suggest the Ubuntu (or your appropriate distribution) user forums. I hope it does work for you, though, and that you get better use out of your Jawbone on Linux.

Disclosure

I’m not sure if the headsets that Cote and James carry (all three of us use them) were bought or are eval units, but mine was purchased for the regular price from the Cingular store down in Brunswick, so no disclosure here.

[1] If you get the message ERROR: Module snd_bt_sco does not exist in /proc/modules, type sudo modprobe snd_bt_sco.
[2] If it doesn’t, and you’re on a Thinkpad, try hitting Fn-F5 to turn bluetooth on, a little green icon should light up.
[3] I played this one.

Colfax During the Double Blizzard

Originally uploaded by sogrady.

Given that the double blizzard is upon us, driving in the snow is going to be an unfortunate reality for pretty much everyone in the Denver metro area for the foreseeable future. As a result, I thought I’d share the most important lessons gleaned from my winter driving education in New England, which included three Maine blizzards.

Rear Wheel Drive

If you have rear wheel drive, you may as well stop reading now: you’re screwed. All the advice and fancy traction control systems in the world can’t make up for the basic physics. So stay home. If you absolutely must go out, the best thing you can do is put sandbags or something similarly heavy (plates from your weight set work nicely, in my experience) in your trunk to try and give the drive wheels better traction. But I’d still recommend staying home.

Go Slow

You’d think this goes would be simple common sense, but in every storm there’s some idiot that’s convinced that because his Hummer has four wheel drive he’s above such behavior. That idiot usually ends up in a ditch somewhere, so take it from me: drive slow. If I can manage this – and I once made it from Boston to Maine going no faster than 20 MPH – you can too.

Avoid Your Brakes

Brakes are your enemy when there’s snow on the ground and should be used as a last resort only. Application of the brakes is likely to cost you what little traction you have and may lock your wheels, so using them is almost always a risk. Use the transmission to throttle your speed, even if you’re unlucky enough to have an automatic. One additional note: if you have ABS, do NOT pump your brakes – the system does that for you. I’m surprised at how many people don’t know that.

Don’t Stop

There are dozens of bad things that can happen when you stop – you could skid, you could be hit by someone who can’t stop, you can lose traction and get stuck, and so on – so do it as little as possible.

Watch Other Drivers

Many experienced snow drivers think they’re in good shape because they’ve been through it before. Regrettably, that’s only part of the equation. You need to be just as concerned about the other folks on the road as you are about your own car – particularly when it comes to intersections, exits and other areas where bad things can happen. As n example, it’s often useful to wait a three or four count before proceeding into an intersection during a snowstorm, so as to avoid being hit by the driver who couldn’t stop to make the light and simply barrels through.

What About Ice?

Avoid it. Snow’s not simple to drive in, but is manageable. Ice is an entirely different matter, and unlike snow four wheel drive means virtually nothing on ice. The same rules apply to driving on ice, but even if you follow them you can easily lose contol.

Stuck?

Rock back and forth and see if you can free yourself. Sometimes you can – like when I got stuck in front of the Irish Hound this very evening – and sometimes you can’t, but blindly applying throttle is likely to only dig you in deeper. And could burn out your clutch, if you drive a stick. If you’re at home, see if you turn up sand, salt, kitty litter, or boards – anything that will a.) give you better traction, b.) melt some of the snow you’re stuck in, or c.) both.

High Beams

Depends on conditions, but you’re probably better off not using them. When there’s decent snow coming down, your high beams are likely to reflect off of the blowing snow and blind you rather than give you better visibility. It’s similar, in that respect, to fog.

That’s all I can think of off the top of my head; feel free to share your tips if you’re so inclined. And to everyone in Denver, be safe out there. The roads weren’t good to begin with, and are getting worse by the minute.

I completely forgot to mention this last week, despite the fact that it’s only pertinent to a distinct minority of my audience, but the fix for suspend after trying AIGLX/Compiz is actually pretty simple.

As I mentioned here, though I managed to get AIGLX/Compiz installed on the machine I did not get them fully up and running, and to add insult to injury I broke suspend on my laptop in doing so. Given that suspend is a feature I was willing to actually switch distributions to get, it’s more than moderately important to me.

Fortunately, the fix was an easy one and didn’t take long to find. For reasons that are unknown to me, the installation process described in the AIGLX/Compiz instructions I used triggered a rewrite of my grub conf file. In the process, it erased a necessary kernel parameter I was passing in. If this has happened to you as well, just do the following (which was copied from here):

gedit /boot/grub/menu.lst And: in your booting kernel option (/boot/grub/menu.lst), add as an option: acpi_sleep=s3_bios

Please note, as the installation instructions will warn you, that mucking around with AIGLX/Compiz and, of course, your boot configuration file does demand a certain level of experience. Not necessarily to make the changes – that part is usually dangerously easy – but to fix and/or back out of them if/when something goes wrong. In other words, don’t say I didn’t warn you.

Just a quick note on some Apache stuff I figured out last night. Those of you who either aren’t technical or aren’t likely to be near Apache might want to skip this one. Ditto for the folks that already know Apache well. Moving along…

The problem I was trying to solve was actually fairly simple. We at RedMonk have multiple domains: redmonk.com, red-monk.com, monkchips.com, peopleoverprocess.com, tecosystems.org and so on. When we were running on top of a managed server, setting up new domains was the province of the managed server interface. Point this domain to this directory, and done.

On our coloed box, however, all the above domains know to do is resolve to our machine. So since our switch, monkchips.com and tecosystems.org, specifically, have been resolving to the basic redmonk.com rather than redmonk.com/jgovernor and redmonk.com/sogrady respectively. With me so far?

In essence, our problem was one of redirection. I needed to redirect incoming requests to another subdirectory on our machine. Seemed pretty simple, particularly because I have some experience with that in the form of Apache .htaccess files. A couple of quick searches led me to conclude that a .htaccess file placed in the root of our webserver that contained the following:

Redirect 301 http://www.tecosystems.org /sogrady

Should turn the redirection trick nicely. You Apache folks in the back of the room snickering at me, pipe down – I’m a confessed Apache noob.

After trying this, I was immediately presented with a very basic problem in that it didn’t work. As in, nothing happened. tecosystems.org continued resolving to redmonk.com as if the .htaccess wasn’t even there.

Back to the forums I went, and I discovered that my .htaccess file was not active for a simple reason: the default Ubuntu Apache config turns it off. The config for an Apache server in Ubuntu reads:

Options FollowSymLinks
AllowOverride None

I’ve bolded the ‘None’ because that was the problem. The none tells Apache that nothing – .htaccess files included – can override the standard configuration. It ignores the .htaccess file, in other words. So that was relatively easily remedied by changing AllowOverride None to AllowOverride All. Patting myself on the back for being so terribly clever, I hopped on a browser, headed to redmonk.com and was presented with an HTTP 500. That’s a problem.

After a bit of trial and error, I discovered that it was in fact the redirection line that was causing the problem. As soon as I removed that, everything was kool & the gang and resolving. Except for the fact, of course, that I still did not our vanity domains working properly.

More research led to the discovery of Apache’s virtual hosts facility. I was loathe to try this because my last experience with the functionality on a Solaris 10 box was while not technically classifiable as a disaster, not terribly fruitful either. But it seemed like the best option available to me.

On Dapper – the version of Ubuntu we’re running as our production server – the Apache file is split up, and the virtual hosts are located in a separate directory – /etc/apache2/sites-enabled, in our case. After looking at the default, and a couple of examples online, I thought that it didn’t look all that difficult, so I created a file (I’d display it, but don’t know how to escape everything in MT) called 001-tecosystems, dropped it in /etc/apache2/sites-enabled, and restarted Apache (/etc/init.d/apache2 restart). What does that do? Go hit www.tecosystems.org and see. As I’ve been told, Apache can be easy to configure – if you know how to do it.

Despite the success in getting the basic domain work done, there are still a couple of problems. The virtual host set up seems to break relative links; the blog templates, for example, were unable to find their CSS files until I provided them an absolute (http://www.redmonk…) path. Normally, this wouldn’t be a problem – as the CSS fix demonstrates – but as covered the other day, Dojo doesn’t like absolute paths so the tooltips over at sogrady.org have broken. Also, I have yet to get subdomains (wiki.redmonk.com) working via a virtual host. When I tried to set up one for the wiki, it loaded wiki.redmonk.com but appended index.php on a dozen or more times. So no joy on that score.

So overall, Apache’s not too difficult to work with. A little obscure and esoteric at times, but manageable. The real question is whether the above – as I’ve contended previously – seems like an opportunity, and if so, for whom? More on that later.

Another excellent day at OSBC, which I’ll get to whenever I get the chance. Given that I’ve got to hop a plane in about a half hour, only to turn around a couple of hours later and head back out that might be a while, but I wanted to relate a couple of quick anecdotes that I discussed last night. As many of you know, I’m a big fan of Josh Schachter’s del.icio.us service, and as of a couple of seconds ago I’ve stored nearly 2500 bookmarks in the service. Despite the service’s popularity, I still have a lot of people question me in just how I use it, and how – precisely – I find it useful. So two quick stories.

1. The first anecdote was related to me by a Friend of RedMonk (who should feel free to reveal themself if they choose). While browsing one day, I came across a link I thought this person and/or their colleagues should be aware of, so I tagged it in del.icio.us with a note to that person. Somewhat coincidentally, this person happened to be demoing del.icio.us to a business associate, and after pulling up one of the keywords I used the link was returned as one of the top few results along with the message to that person. In other words, it’s a quick, easy way to collaborate in a more permanent fashion than email.

2. More interesting, perhaps, was how I used to it to help prepare for my Zend talk. Knowing that I was going to be talking about simplicity and less code, I tagged items as I ran across them well in advance. Then, when it came time to put together the presentation, I simplicity visited my del.icio.us links for simplicity and less code, and voila – I had everything from the AK-47 to the bug story all neatly linked in one place. Was a hell of a lot easier than a.) trying to remember what I wanted to talk about and b.) trying to hunt it down again. Plus, it’s far more natural and sustainable a process than taking presentation specific notes.

So while these examples are relatively simple, I thought they might give a few of you an indication of just how you might be able to use del.icio.us. Enjoy.

This is an entry that’s probably not of interest to most of the folks reading this, as a high proportion of you run your own blogs, but if either you or your coworkers, friends, or family were desperately seeking a resource on how to start blogging, you might point them here.

The link is to a HowTo I just did to compliment an upcoming client session. I picked Blogger simply because it’s easy (no, Google’s not a client), but if anyone feels motivated enough to do something similar for TypePad, MSN Spaces or the like – knock yourself out.

And of course, given that it’s a wiki, you should feel free to correct and/or correct the existing entry. Hope it’s helpful.

Just a few really cool things you can do with some of the free web services (small w – not SOAP/XML/etc stuff, though that is sometimes involved) out there. And just as a disclaimer, I’m neither associated with nor cover any of the services mentioned below. I just happen to think they’re cool.

Use Social Bookmarking:

Many of you have probably read or heard of social bookmarking, or maybe you haven’t. Either way, many – most, actually – of the folks I speak to don’t see the value. So here are a few ways you can use http://del.icio.us:

1. See what the web is buzzing about: http://del.icio.us/popular
2. Research a topic – for example, privacy. Simply take your term and replace privacy: http://del.icio.us/tag/privacy
3. See what people who link to something you like, link to. After we both linked to Tomboy, I found that atariboy linked to things I was interested in: http://del.icio.us/atariboy. He found the link I describe in the next one. This works much better for niche or rare links.
4. Visualize the metadata topics you’re covering (see here)
5. Find new things via a simple tag. I subscribe, for example, to http://del.icio.us/tag/useful and often discover – go figure – useful things

Alert Feeds in Bloglines:

Wish I could take credit for this one, but one of my readers (Vaibhav) put two and two together where I failed to. Basically you can sign up for a Google New Alert, as I outlined here. But instead of routing it to your regular email, you can push them over to Bloglines. How? Simply scroll to the bottom of your My Feeds tab in Bloglines, and hit “Create Email Subscription”. Then simply sign up to the alert with the email provided in your Bloglines email subscription. All of a sudden, your custom news alerts will show up in your Bloglines reader, rather than cluttering up your email box.

Use Social Photo Software:

Like social bookmarking, Flickr is a shared metadata service, but applied to photos rather than bookmarks. Sounds cool, but most people are sort of short on ideas of how to use it. I’m no expert – just getting use to it myself, but here’s a few ideas:

1. Subscribe to an RSS feed of your favorite sports team. Then, whenever someone takes a photo and puts it in Flickr with that metadata, you’ll see it. The Red Sox feed is as follows: http://www.flickr.com/services/feeds/photos_public.gne?tags=redsox&format=rss_200
2. Search the available tags here, and subscribe to a tag that interests you. Maybe it’s Ireland, maybe it’s cathedrals, maybe it’s dogs. Whatever. Point is, you can subscribe to it.
3. Create a private group, and share photos within it. Maybe it’s your group of friends, maybe it’s your project team.
4. Check out my photos – only a couple in there, but still: http://www.flickr.com/photos/sog/
5. Add photos to your blog here

Use Feedburner for Your Blog feed:

I was sort of put off by the idea of Feedburner at first, which is essentially that when people’s readers pick up your feed, you send them over to Feedburner to pick up an optimized version. It seemed cool because I’d get a better read of how many people were reading the blog, and what they were clicking on, but apart from that I didn’t see the value for the hassle. I was wrong. I’m now using Feedburner to splice in my del.icio.us links, meaning that those people picking up my blog feed are now able to see the links I’m reading every day as well. I could do photos as well – we’ll see about that. But it’s a very nice service – and the hassle? Couldn’t be easier. Anyone hosting on Apache just has to drop in a one line text file and you’re good to go.