Dang, that’s small. Continue reading Raspberry Pi — credit-card sized Linux
I think Dropbox is a neat idea. It makes file sharing between my several computers and optionally my friends’ computers — of course only that content which I am legally entitled to distribute — really, really easy. It mostly just works, on Linux, Mac, and Windows. Not to mention my and Sarah’s iPhone. It’s great for keeping your stuff available in multiple places and sharing certain files with certain users, but not everyone (like from inside the dressing room when you weren’t sure which dirndl was the winner).
Dropbox needs to get paid for any significant storage capacity, however, and I’ve got a very nice virtual computer in a high-powered data center somewhere mostly sitting around idle, waiting to do some boring stuff like make this content hit your eyeballs. And even Dropbox is not immune to security flaws and breaches.
Why not leverage that extra storage and bandwidth capacity by storing stuff for safe-keeping or sharing more than what you get for free from Dropbox on your own server via SparkleShare? It’s free (beer, gratis) and free (freedom, libre) and doesn’t limit you to a certain number of users/clients/files/bytes of storage. I’ve been playing around with it for about a week, and I like what I see so far.
It’s available for several platforms:
- Mac OS X
… and iOS and Windows clients are in the works. Setting up a server on Linux was pretty easy if you are comfortable with package managers and user administration (and if you have a Linux machine running server software of any kind at your disposal, you better be). SparkleShare runs on top of Git and OpenSSH, so you need to have those packages installed and running. I’m not much of a coder, especially not in a group setting, so I had no need for Git prior to setting up SparkleShare.
The Mac client Just Works™. The Linux client was more of a challenge for me. The Ubuntu package was pretty far out of date, so I gave it a shot building from source. That was kind of an adventure, but in the end I got it running by combining some files from the 0.8 version and some from the 0.8.1 version. When it comes to compiling software, I’m mostly just fumbling around in the dark, so I’m sure it took me much longer than someone who actually has a clue.
So, what’s keeping SparkleShare from taking over the world of redundant backups and file sharing? A few things:
- quick setup à la Dropbox
- a web interface for administration (also supposedly in the works)
- the aforementioned Windows & iOS clients
- being based on Git, you need about double the storage space that your files themselves take up
That last one there confused me, since I’m not a Git user. According to Wikipedia, Git stores all the history of changes to the files being shared in every local copy. This is great if you want to look back in time at previous versions of the files, but not so great if you’re low on storage space already. SparkleShare’s programmers are looking into opening their software up to other methods of replication as well. With those, you might lose the versioning history but save on disk space.
I think it’ll really take off when it’s as easy to install on the client side as Dropbox. The only downside at the moment is that I don’t have enough friends on Mac/Linux computers with whom I regularly share files to really give it a proper field test, but it’s working great on our LAN here at the apartment. I’m using it to keep a couple hundred MB of my favorite desktop wallpaper photos in sync between a couple of Macs, and that’s working out great.
- 0c45:613c Microdia PC Camera (SN9C120)
to mostly no avail. I kept finding hits and solution suggestions for older operating systems, or newer cameras. But I pressed on, because I knew this camera was working with applications like the aptly named Cheese (whence the image in this post came), Kopete, and even Google+ Hangouts seemed to work just fine with this cheapo camera (well, aside from its cheapo image quality, that is).
Sooner or later along the way, I hit upon this post by Hitesh Sarda which gave me a hint about starting Skype by setting some environment variables. I tried what he suggested, and that didn’t work for me. But I noticed that in the same folder as the shared object he mentioned, there was another one available.
ov511-decomp ov518-decomp v4l1compat.so v4l2convert.so
And that v4l2convert.so shared object file does something to make my camera work with Skype. So with a simple shell script, I can start Skype and get the video working:
env LD_PRELOAD=/usr/lib/libv4l/v4l2convert.so skype
I suppose most people would have simply upgraded to a newer/better webcam…but not me. Skype needs to catch up with other software companies supporting more varied hardware via open-source platforms out of the box, otherwise they’re going to end up just like Microsoft Internet Explo — oh…uh, never mind.
Haven’t geeked out here in a while!
This morning I put Kubuntu 10.10 onto my Mac mini’s hardware (in parallel to OS X 10.6) sort of for a lark and sort of because I like the KDE desktop so much. It went surprisingly smoothly and painlessly, despite my hodgepodge of cannibalized hardware (but hey, that’s what a Mac mini is for, right?).
I had no trouble getting all the basics to work: external hard drives readable and usable, WLAN was a snap, even got Kdropbox (though it’s not called that anymore) going quickly and painlessly.
Except for Skype. I downloaded the latest .deb package from skype.com and the audio output stuff seemed to be working just fine. But not the audio input from my USB microphone. Couldn’t figure that out. I did some googling, but mostly found references to Skype beta versions and hacked up wrappers from years ago, like version 1.2 or 1.3. Didn’t really seem appropriate anymore. There was a hint in the Skype Sound Devices options about PulseAudio and messing the with the settings outside of the Skype software.
So I kept digging in the KDE System Settings dialog. First weird thing I noticed: no obvious “Sound” or “Audio” control panels? Hmm, what about this Multimedia setting? I looked in there more than once before I realized that I need to take a look at “Phonon”.
Where do you have to put in a username and password in your daily computer geekery? Here’s what it looks like for me.
my laptop running Windows • our crappy corporate email client • our crappy corporate travel provider • the software that controls the phone on my desk • lots of other programs not smart enough (or not allowed) to authenticate me based on other methods
our Mac at home • our Linux desktop at home • our Linux laptop • our email provider • our bank • Skype • Paypal • Amazon.com • Amazon.de • iTunes • Twitter • dozens, if not hundreds more
I imagine your situation is similar. With the personal stuff, you really should not be using the same passwords at multiple websites. Just one site being sloppy about security and getting breached by hackers is enough for them to send email in your name and steal money or service from you — look what happened to usernames and passwords recently at Gawker Media. You probably know someone whose account got hacked with real-world financial implications — I know two people to whom this happened in 2010 (and a third who got hacked but apparently didn’t lose any money). It happens all the time.
You and I both know you this is not a safe practice. But what can you do about it? With so many usernames and passwords in your daily life, the natural inclination is to stick to just a few username/password pairs and reuse them entirely or perhaps modify them slightly. Writing down passwords and usernames onto paper might be OK at your home, I guess, but that means you need to carry that piece of paper with you out into the world if you are going to do any sort of mobile computing. Writing those usernames and passwords onto paper at the office is a terrible idea; don’t ever let your IT people know that you do it.
Instead, you can use Password Safe on Windows or a compatible program like Password Gorilla on Windows / Mac / Linux — and even on your iPhone or iPod Touch via the PasswordVault app. Instead of those hundreds of username/password combinations to remember (or look up), you only have to know one password to get into your “safe.” From there, you can copy usernames and passwords with the mouse (and keyboard shortcuts) from the “safe” into whatever application is requesting your credentials. Password Safe can randomly generate passwords for you based on policies you define: minimum password length, exclusion of easily mistaken characters (like zeroes/O’s or ones/L’s), inclusion of punctuation characters, etc. Lots of cryptologically sound practices there. “But how will I ever remember those randomly-generated passwords?” you ask? Well, you won’t. You’ll have to remember the one password to get you into the “safe” and the application will remember the rest for you.
I keep my “safe” file updated on my Windows computer, and then synchronize that periodically to my Mac and Linux machines via Dropbox. From my Mac, it synchronizes into my iPod touch. This means I am carrying that piece of paper with all the sensitive info on it around with me after all, but in electronic and encrypted form: I still have to enter the password to open the “safe” on all those computers/devices in order to get a glimpse of the content.
But hey, I can remember one password pretty easily, especially if it virtually eliminates the chances of someone stealing my purchased Skype-out credits or impersonating me via a hacked Gmail account.
Sounds creepy, doesn’t it? Like I’m going to corner you at a party and explain how
Beautyvision is more than just a franchise opportunity… It’s a way of looking at the world with the total absence of fear.
It’s not like that.
We got pretty tired of picking hotels and rental apartments based in part on their advertised internet connectivity. Which pretty much always is more hassle than it’s worth. Therefore, when I heard about O2‘s “dayflat pack” rate of €3,50/day or €25/month on a prepaid surfstick, I thought it was a great idea. We have a pretty reliable internet connection at home, and don’t need a mobile internet contract because we’re not usually out on the road. But when we
“Whew!” I thought to myself. “This sounds great!”
Reading a little closer into the system requirements, I saw that linux — of course — isn’t supported. Googling a bit, I found you can’t even fake it via Wine.
A little more googling revealed stuff about usb_modeswitcher and umtsmon — programs which enable your linux box (or netbook in this case) to manipuate your phone’s (or surfsticks’s) SIM card while it’s plugged into your USB port or PCMCIA slot. I noticed all the posts on this topic were about two years old, and apparently largely irrelevant now that Ubuntu supports all those things more or less out-of-the box.
But I’m on Kubuntu, which always seems to lag behind Ubuntu in terms of networking, and was still struggling until I coughed up for the Huawei E5 mobile WLAN router. This cute little device will let me and my four favorite WiFi capable devices surf on the cellular signal from that O2 pre-paid card: Sarah’s phone, our little netbook, my iPod touch, plus two other WLAN-capable devices (perhaps those of travel buddies), all simultaneously. Best part: somehow it knew all the settings like APN and dial-up number (huh? dial-up?) or, it read all that stuff off the SIM card…not sure which. Kubuntu and umtsmon needed me to tell them those things (and I was clueless as to what to put in there).
There is at least one caveat here: the easiest way to tell O2 you want to cash in some of your prepaid credit in the form of a dayflat pass is to use the software that came installed on the surfstick’s tiny little drive. That requires Windows (which I only use at work, and not at home) or a Macintosh (OK, got one of those). But my Mac is not a mobile device. How then, can I activate a dayflat pack while underway armed with only the SIM card (and surfstick) and no Windows/Mac OS with me?
The secret is that you CAN purchase a daily or monthly pack without the O2 “Mobile Partner” software. You just put that O2 prepaid SIM card into any GSM phone, boot it, and then dial
*104#. I’m not sure what this is called in English, but the text that appeared on my phone said something like “service command” (I think was “Dienstbefehl“). From there you can enter numbers on your phone’s keypad to purchase a dayflat or monthly pack out of your pre-paid account’s available credit. Then pop that SIM card back into your surfstick or portable WiFi router and you’re good.
My venerable old vprMatrix (I think it was a Best Buy house brand) computer has gone through a power supply and video card and I’ve beefed up the hard drives on it a couple times. Now I think it might be time for its second video card. I upgraded it from Kubuntu 8.04 (Hardy Heron) with the intention of going straight to Kubuntu 10.04 “Lucid Lynx” LTS, but alas, the 8.04 “Hardy Heron” release of Kubuntu was not deemed an “LTS” release, so I have to hop, skip and jump my way around here a bit to fully modernize the OS on that box. If I even get all the way there.
The upgrade seemed to go OK to 9.10, but once I rebooted into Kubuntu 9.10, I noticed one thing immediately: the snazzy new (to me) KDE4 environment was not drawing the windows properly – parts of most (but not all windows) were displaying ostensibly correctly, but the title bar of every window was garbled with a repeating pattern of lines — some vertical, some horizontal. So much so, that I couldn’t read the titles of any of the windows I clicked on. I had to rely on my old friends, the keyboard shortcuts, to call up a console session and turn the machine off via command line. I googled around a bit and found this: Window corruption with older ATI graphics cards
The tips there fixed my window display problems, but I wonder if I should just upgrade the video card anyway. Scrolling seems clunky; 32MB of RAM on the graphics card just might not cut it anymore. We’ll see how it looks when I get to 10.04 (another 1500 files have to be downloaded and configured and installed and …).