This is where online backup services come in handy. And there’s plenty of them, which makes it hard to decide who to entrust with your sacred files. Fortunately, some people already did the work and made some great charts that enlist all the benefits and downsides of the individual services.

What I was (and still am) primarily looking for is:

1. unlimited free online storage
2. no filesize limitations
3. client software that does incremental backups
4. webinterface to access my data from anywhere

Point 1 of my wishlist seems unreasonable and ridiculous at first but there’s indeed one service (Mediafire) that will give you unlimited free storage space. The downside is that the filesize is limited to 100 MB and there’s no client software, making the service more of an “upload and forget” type’o thing.

Crossing “unlimited storage space” off of my wishlist, I went for the “next best thing” and looked into Oosah, which gives you 1 TB of free storage space. However, they limit the filetype of files you can upload/backup to “media files” only and also lack support for incremental backups.

Next in line is ADrive, which actually met all my requirements from above and therefore seemed like it hit the bulls’s eye. However, their “backup software” is merely a modfied FTP client that WILL DO incremental backups, but is terribly slow in doing so. I could live with the speed, but on top of that the software doesn’t seem to support UTF-8, which results in some of my file names (those with special/german characters in them) being truncated, cutting off the extensions and thus making them directories. *dropped*

I could go on an on about how I tried Humyo (30 GB with sign-up; 25 GB must be media files), Windows Live Mesh (5 GB for free) or Orbitfiles (6 GB for free), but to make a long story short….none of them met my demands.

Frustrated I dug through the comments of the lifehacker article, and was about to put down some money on Jungle Disk (a per-GB-per-month backup service hosted by Amazon’s S3 service), which seems like a good, reliable deal if you’re willing to pay for backups. But then again, I don’t….at least yet

Instead, I went with an old friend of mine: Wuala. While their interface (written in Java) is indeed one of the worst out there and they only provide you with 1 GB free online storage when you first sign up, you can trade in some of your harddisk’s GBs against online storage space. If you provide Wuala with e.g. 10 GB of your local harddisk (which is then used to store encrypted data of other Wuala users) and your computer is online at an average of 80% of the day, you will recieve 8 GB of online storage…simple as that.
On top of that, Wuala provides filesystem integration (maps your wuala account as a network drive), which enables me to use any backup tool (ViceVersa in my case) that I would normally use for local backups. Wuala will cache the files I copy onto the “Wuala drive” and successively upload them. You could even go as far as storing all your files on the Wuala drive and working with them as if they were local.
They also have a (Java-) webinterface that enables me to access my files from any computer.

I just released the source code of iTunesSnarl. It is available via Google Code. So if anyone out there could please fix the cover issues, I’d be more than happy

Check it out here: http://code.google.com/p/itunessnarl/

Version 1.1, released 2008/09/04

• The current song’s title is now displayed in the title of the snarl window.
• The snarl window now also lists the artist’s album title (most of the time….don’t know why it bails out sometimes).
• Removed Snarl_CMD.exe and any dependencies to it.
• Plugin now registers with Snarl and popups can be disabled via Snarl.

I wish I could tell you that now the album cover is also displayed in the Snarl window, but as I said above, that proofs to be a little more difficult than expected. Any help is appreciated!

The installer can be downloaded here.

———————————————————————————–

Update: Go here for version 1.1

———————————————————————————–

Over the last weeks I’ve become a huge fan of Snarl, which is a Growl-like notfication system for Windows. With the appropriate plugin, applications like Firefox or Miranda can display user messages through Snarl, which only opens small “bubble” message boxes that close itself after a certain period of time. This way the desktop isn’t cluttered with message windows that the user has to close by hand.

So far, the number of extensions/plugins that are available for Snarl is still small, but with the software below I’m doing my part to increase it

Since the only plugin for iTunes, iTuny, is more of a Launchy plugin than a Snarl plugin, and since I couldn’t get its Snarl component to work (maybe I didn’t try hard enough), I decided to give the iTunes SDK a try and developed my own plugin. You can download the installer below. When running the setup, the plugin will install itself to $Program Files$\iTunes\Plugins\iTunesSnarl\ from where iTunes will load it automatically whenever you launch it. So no need to start the plugin seperately (of course Snarl needs to be running). Just fire up iTunes and that’s it. From now on you will get a small bubble message showing you the current artist and track title whenever you skip from one track to another.

Download it here

P.S.: I borrowed the Snarl_CMD.exe from the iTuny sources. f0vela, I hope you don’t mind!

Unter http://www.nichtlustig.de gibt es wunderschöne Comics über depressive Elefanten, ängstliche Zebras, kreative Lemminge, sowie viele andere skurrile Alltags- und Nicht-Alltagssituationen. Absolut sehenswert!!!

After playing around with the code for a while, I got some weird exceptions thrown at me, e.g. “A first chance exception of type ‘System.Runtime.InteropServices.COMException’ occurred in PresentationCore.dll“. Now, the only thing I had changed was letting the tagging routine run in a seperate thread:

[STAThread]
static void Main()
{
Thread t = new Thread(writeMetadata);
t.Start();
Application.Run(new Form1());
}
static void writeMetadata()
{
// write the metadata to the image here
}


After searching the web for a while I found an article by Aaron who had run into similar problems. Apparently, WPF components cause some trouble when they’re executed outside a thread that doesn’t have its apartment state set to STA. Fortunately, Robert was kind enough to point me in the right direction. All you have to do is set the apartment state right and off you go with WPF components all over your threads Thus, the Main() routine from above changes as follows:

[STAThread]
static void Main()
{
Thread t = new Thread(writeMetadata);
t.SetApartmentState(ApartmentState.STA);
t.Start();
Application.Run(new Form1());
}

That’s all there is to it!

After searching for a solution for days, I stumbled upon AppUpdater, which is a free .NET component that “was written using the .NET Framework and enables you to make your application auto-updatable simply by dropping the component into your existing .NET applications and setting a few properties (ex. where to get updates from)“. And indeed, it’s as easy as it sounds. Drag-and-drop the component on your application’s main form, configure the AppStart.config file and start your application by using the provided AppStart.exe. Then put an updated version into the according folder on your server and the AppUpdater component will take care of the rest. (if you cannot follow, read the AppUpdater article first

Although step 5 of the walkthrough says that any Web-DAV enabled web server should suffice, I ran into trouble using an apache web server with the dav and dav-fs modules loaded. Apparently, you need an IIS, since it interprets DAV verbs a little different than its apache counterpart. While both the CheckForFileUpdate() and the UpdateFile() functions from the AppUpdater’s class “WebFileLoader.cs” work as expected, the client throws an exception in GetDirectoryContents(). It seems that the apache dav module has some problems with the iscollection and displayname tags of the request string since it returns the following response (e.g. for the file appupdater.dll):


<D:response xmlns:lp1=\"DAV:\" xmlns:lp2=\"http://apache.org/dav/props/\" xmlns:g0=\"DAV:\">
<D:href>/update/1.0.0.1/appupdater.dll</D:href>
<D:propstat>
<D:prop><lp1:getlastmodified>Sat, 12 Apr 2008 00:30:16 GMT</lp1:getlastmodified></D:prop>
<D:status>HTTP/1.1 200 OK</D:status>
</D:propstat>
<D:propstat>
<D:prop>
<g0:displayname />
<g0:iscollection />
</D:prop>
<D:status>HTTP/1.1 404 Not Found</D:status>
</D:propstat>
</D:response>

getlastmodified returns the right timestamp, but both displayname and iscollection return a “404 Not Found” error. Now, I could’ve gotten into the DAV protocol, put my web server into debugging mode, TCP-dumped a couple of DAV requests and tried to build a proper request string, but then again there is actually no need to employ DAV for simple tasks like this. The simple HTTP protocol totally suffices in this situation. You just need to go a slightly different way:-) As initially implemented, the AppUpdater connects to the web server in the GetDirectoryContents() function and browses the directory to find and copy all the contained files to the client. “Directory Browsing” is a DAV feature, so we’ll have to improvise here.

Let’s say I just released version 2.0.0.0 of my application (Former version was e.g. 1.0.0.0). According to step 5 of the article, I simply need to create a new folder named “2.0.0.0″ in my server’s root update directory and edit the UpdateVersion.xml file to point to this newly created dir. Instead of uploading all files from our release folder, we’ll zip them up and upload the resulting zip-file into the server’s “2.0.0.0″ folder. That way our client only needs to download one file (via HTTP) and isn’t dependent upon the “Directory Browsing” feature. After downloading the zip-file we’ll just unpack it into the temp directory that the regular AppUpdater would have stored the downloaded files in and let it do its magic from there on. It’s as simple as that, plus you only need to zip up and upload those files, which have changed between releases. The AppUpdater will copy all files, which it finds in the 1.0.0.0 but cannot find in the 2.0.0.0 folder to the latter folder. Unfortunately, this little hack doesn’t allow for auto-downloading of missing assemblies anymore (setting the AutoFileLoad option of the AppUpdater component to true), but then again users can just download the installer and re-install the app, if they accidently deleted any of the files in the app’s root folder.

Now to get going and to enable the AppUpdater to work with your apache web server (without Web-DAV) you actually only need to modify one file. First download the the .NET Application Updater Component package, which includes the AppUpdater’s source code. Afterwards edit the file “WebFileLoader.cs”. Since the GetDirectoryContents() function caused us so many headaches and we won’t need it anymore you can delete it entirely. Afterwards, replace the code of the function CopyDirectory with the following (most of which is copied from the UpdateFile function):


Resource currentResource;
int FileCount = 0;
string newFilePath;
if (!Directory.Exists(filePath))
Directory.CreateDirectory(filePath);
HttpWebResponse Response;
HttpWebRequest Request = (HttpWebRequest)HttpWebRequest.Create(url);
Request.Credentials = CredentialCache.DefaultCredentials;
try
{
Response = (HttpWebResponse)Request.GetResponse();
}
catch (WebException e)
{
if (e.Response == null)
{
Debug.WriteLine("Error accessing Url " + url);
throw;
}
HttpWebResponse errorResponse = (HttpWebResponse)e.Response;
if (errorResponse.StatusCode == HttpStatusCode.NotModified)
{
e.Response.Close();
return 0;
}
else
{
e.Response.Close();
Debug.WriteLine("Error accessing Url " + url);
throw;
}
}
Stream respStream = null;
string newPath = "";
try
{
respStream = Response.GetResponseStream();
newPath = filePath + Path.GetFileName((new Uri(url)).LocalPath);
CopyStreamToDisk(respStream, newPath);
DateTime d = System.Convert.ToDateTime(Response.GetResponseHeader("Last-Modified"));
File.SetLastWriteTime(newPath, d);
}
catch (Exception)
{
Debug.WriteLine("APPMANAGER: Error writing to: " + filePath);
throw;
}
finally
{
if (respStream != null)
respStream.Close();
if (Response != null)
Response.Close();
}
// unzip files
Shell32.ShellClass sc = new Shell32.ShellClass();
Shell32.Folder srcFlder = sc.NameSpace(newPath);
Shell32.Folder destFlder = sc.NameSpace(filePath);
Shell32.FolderItems items = srcFlder.Items();
destFlder.CopyHere(items, 20);
File.Delete(newPath);
return 1;

(Sorry for the messy code. I couldn’t get Wordpress to indent it properly:-( In the last 7 lines we employ the Shell32 COM library to extract our zip file. To make this work, you need to add the correct reference. In your project, click on “Add Reference”, then on the “COM” tab in the opening dialog and then search for “Microsoft Shell Controls And Automation”. That’s the lib you want to add. Finally compile it and add the resulting .dlls to your project, although I would generally recommend adding the entire AppUpdater project to your solution and referencing it.

Before we can test our new application, we need to create a new directory on our server (let’s assume the example from above and name it 2.0.0.0). Then go through the article’s steps 1-4 and do everything as told, build your application, zip up the new files (to e.g. app.zip) and upload the zip file to the 2.0.0.0 directory on the server. Afterwards, edit the UpdateVersion.xml file as follows:

<VersionConfig>
<AvailableVersion>2.0.0.0</AvailableVersion>
<ApplicationUrl>http://your.web.server/your_update_dir/2.0.0.0/app.zip</ApplicationUrl>
</VersionConfig>

It’s very important to mention here that with the modifications you made, you now need to give the FULL PATH to your zip file within the ApplicationUrl tags, not just the directory as before.

Last but not least, fire up your client and let it automatically update itself!

That’s it! Hope it works for you.

Today I released an updated version of my Skype plugin “Contact Alarm“. It introduces a new setting, which will let you replace the default alarm sound with one of your own .wav files. In addition, the license bug was (hopefully and finally) resolved.

So if you experience any problems, or if you have any questions or suggestions for the next release, feel free to let me know in the comments or drop me an e-mail.

Get it from Skype Extras

If you consider a donation (which would make me happy:-)), you can do so with the button below.

This work is licensed under a Creative Commons Attribution-No Derivative Works 2.0 Germany License.

In der letzten Zeit habe ich viel über virales Marketing gelesen und gehört, speziell von einem Freund, der in diesem Bereich seine Diplomarbeit schreibt. Heute morgen ist diese neuartige Vermarkungswoge bis in die Wrangelstraße, genauer gesagt in meinen Briefkasten geschlagen. Dort fand sich ein gefütterter Umschlag, der nach dem Öffnen ein schmales Holzkästchen zu Tage förderte. Selbiges enthielt neben einer Lochkarte, welche auf der Rückseite die Beschriftung “THE FINAL MILL INC.”/”SERIES AAB” trug, einen Zettel mit folgender Info: “To execute these commandments you’ll have to find the properly gifted fellows.” Der Poststempel besagt, dass der Umschlag in Frankfurt aufgegeben wurde.

Torsten, der sich so schön hat anstecken lassen(:-), und meine Wenigkeit haben haben daraufhin versucht, zumindest die Lochkarte auszulesen…wie es scheint jedoch erfolglos. Was wir haben ist: “UMAKZZLNBZWHTWNYOLVIFKPRGQCIPNJIQTRRFRJMIGXGLOGEVIEJWNNCRZZCGKQW”. Wenn man die einzelnen characters jeweils mit einem offset versieht, kommt leider auch nichts vernünftiges dabei heraus.

Nach einer erneuten Google-Suche bin ich auf Thomas gestoßen, der scheinbar ein ähnliches Paket bekommen hat. Nach seinen Fotos zu urteilen, ist seine Lochkarte jedoch nicht identisch mit meiner, weshalb ich mittlerweile davon ausgehe, dass darauf kein wirklicher Hinweis zu finden ist. Vielleicht gibt es ja irgendwo noch “gifted fellows”, denen man ergänzende Infos zukommen lassen hat. Bis dahin tue ich, was von mir erwartet wird, mache diesen Blog-Eintrag und gieße Wasser auf die Mühlen der Werbeindustrie. Ihr seit super!

Update 1: Neben Thomas und mir haben scheinbar auch Robert und Mariusz Post bekommen.

Update 2: Hier nochmal ein link zu einem höher aufgelösten Scan meiner Lochkarte. @Stefan: ich hoffe, Du kannst damit was anfangen!

Update 3: Auf http://www.wiki.argr.de/index.php5?title=Final_Mill_-_Trail werden Infos gesammelt.