DasGenie:!Scrap

My first S.M.A.R.T. failure

Thu, 24 Jan 2008 00:32:54 CET

My bad luck with hardware continues to amaze me. Bought in September last year, my Western Digital 250 GB latop drive has a S.M.A.R.T. failure. A first for me, normally my disks fail first and then show a S.M.A.R.T. error. How did I notice? I accidentally openend disk Utility and my disk showed up red. What I am wondering is why apple doesn’t show me the S.M.A.R.T. status more prominently somewhere if it fails?

If your disk has a S.M.A.R.T. failure and you are on a Mac then you can get more information about the failure using SMART Utility – which shows me this:

Which IMHO means that this time I could be lucky and no data has been damaged. So here I go and order a replacement from the Western Digital website which much to my suprise offers a variant where I get the new disk first, and have to send the broken one back in a 30 day time window. So if my disk holds up until then I can safely clone it using disk utility. If not, I have my time machine backup and will see how well it will restore my old state. Wish me luck!

P.S.: I just found SMART Reporter which actually checks for the S.M.A.R.T. status all the time, and shows up an alert and red hard drive if it detects a failure. It is freeware, and has just been added to my startup items.

Bugs, Child Windows and Spaces

Tue, 22 Jan 2008 12:04:35 CET

During my recent SubEthaEdit bugfixing efforts I came along some pretty annoying Cocoa Child Window issues.

First a little preamble, why did SubEthaEdit use child windows – this could be filed under “it was a bad idea anyways”. I got the idea searching for a good and safe way of displaying the little SSL-lock icon in the upper right of the window if a connection is SSL-encrypted. Safe meaning a way that wouldn’t break if something fundamental in the way how the window title is displayed would change in the future. Since most of the displaying in the titlebar window area is private API, the results I found when looking around did use some mean subclassing. I didn’t want to do that, so I thought having a tiny transparent child window on top of my window would be a very safe thing to do. So I did. And I was wrong.

Bug number one: Spaces breaks

This seems to be a known issue, when I filed it it almost instantly got duplicated to this one: Child windows and Spaces [rdar://5593261] .

If you have a window with a child window it behaves strangely on spaces. If you open spaces and drag such a window to another screen, then it does not move to the point you dragged it, but to the point it was on the previous screen – much like when you press shift or apple while dragging windows in the spaces view. That in itself wouldn’t be a major problem, but here you go: if you have multiple windows with child windows, all of them gather together on the screen you activate after moving a window around. Hurray, no more spaces for you puny little app :(. So if your app behaves this way, that might be the problem. This was not easy to find out, I found it by chance after making sure that both the main window and child window did have the same collectionBehavior. (P.S.: if you want to put a view in the titlebar, simply ask a standard button for its superview and frame and add some view relative to that. Thanks to Panic for this idea which sounds quite safe too.)

id superview = [[[self window] standardWindowButton:NSWindowToolbarButton] superview];

Bug number two: Your app crashes on quit in certain situations

We had this nasty little bug that every now and then crashed SubEthaEdit on quit. Quite an annoying habit, but we did not find a way to reproduce it. Thankfully amongst the many bug reporters who sent us the crash log via the built in reporter, there was one that actually found out a way to reproduce it. The crash only happens if you quit the application while it is in background. Then the child window is somehow overreleased. You can see the setup code of my sample NSDocument here:

- (void)windowControllerDidLoadNib:(NSWindowController *) aController

{

    [super windowControllerDidLoadNib:aController];

    // Add any code here that needs to be executed once the windowController has loaded the document's window.

    {

        // add a transparent child window to display a lock next to the toolbar button

        NSImage *lockImage = [NSImage imageNamed:NSImageNameInfo];

        NSRect frame=NSZeroRect;

        frame.size = [lockImage size];

        frame.origin = [[aController window] frame].origin;

        frame.origin.x += 100;

        NSWindow *childWindow = [[NSWindow alloc] initWithContentRect:frame styleMask:NSBorderlessWindowMask backing:NSBackingStoreBuffered defer:NO];

        [childWindow setIgnoresMouseEvents:YES];

        NSImageView *imageView = [[NSImageView alloc] initWithFrame:frame];

        [imageView setImage:lockImage];

        [childWindow setContentView:imageView];

        [imageView release];

        [childWindow setOpaque:NO];

        [childWindow setBackgroundColor:[NSColor clearColor]];

        [[aController window] addChildWindow:childWindow ordered:NSWindowAbove];

        [childWindow release];

        NSLog(@"%s %@ %@ %@",__FUNCTION__, childWindow, imageView, lockImage);

    }

}

This is reproducible both in Leopard and in Tiger and a very bad thing. Not really grasping what goes wrong exactly in that case (if you close the windows or quit normally everything is fine, hence no general memory management problem – update: look at the bottom of the article to see that I was wrong nevertheless) I fixed this one in the 3.0.2 release by changing the lifecycle of the child window. I did so by putting the child window into an instance variable of my window controller subclass, and autorelease it on the dealloc method of it.

What I did not know then was that this didn’t fix it on Tiger in a special occasion: If on Tiger you quit SubEthaEdit while it was not active and had unsaved changes, and then without activating SubEthaEdit first directly click on the “don’t save” button. On PPC-Tiger that is. So I revisited the bug after a bug reporter finally added this magic information of reproducibility. What I did to fix it first was retain all child windows in the AppControllers applicationWillTerminate: delegate method. Which interestingly enough needed me to make an extra class for theses windows to identifiy them, because in the applicationWillTerminate: method the child windows were still alive – the parent windows weren’t. However I finally got rid of all of this by using a different and better method of adding the SSL-ock icon to the window, but some of you that might not be in the luxury of removing their child windows might find my trip valuable.

This one hasn’t been made a duplicate yet and is quite annoying. So if you have the same problem, report a bug referencing this one: Child window overrelease [rdar://5686435] .

Sample Code

If you want to reproduce it, here is my ChildWindow sample I wrote for the bug report. If you want to reproduce the last part of bug number two you have to add an NSWindowController subclass to the sample. Thanks for listening.

Addendum isReleasedWhenClosed

I have just been informed by the nice guys at Apple that programatically created windows default to a isReleasedWhenClosed state of YES which explains the overrelease. However why the heck it doesn’t crash when a window is closed does still not seem right to me. And what that means is that if I add the window without autoreleasing or releasing it beforehand that my app leaks windows. Which I verfied using ObjectAlloc – so it is still a bug, but a different one. If you want to do it right you have to setReleaesedWhenClosed:NO on the child window before you add it and everything is fine.

SubEthaEdit 3.0.2 released today

Wed, 5 Dec 2007 01:21:56 CET

This is “just” a point release – but we put in much effort to find many little annoying crashes which hid in special cases. There is nothing more unsatisfying than a crashing product. We know that and are very thankful of your bug reports which let us find out and fix the ones that are harder to reproduce. We especially want to thank the ones of you that take the time and reduce the problem to something that is reproducible.

So go ahead and grab the update via “SubEthaEdit -> Check for updates…” or directly via our Website . For your feature requests or bug reports please take the time and enter them into our Bugtracker – or just give us feedback via our Feedback Form – we value your input and feedback very much.

Comments in the New Pages...

Tue, 7 Aug 2007 22:58:51 CEST

...really mark badly, and print even worse. See for yourself:

All of the lines are off, there are white stripes and some Text is overlapped with markers…

Nice viral marketing...

Fri, 27 Jul 2007 14:17:16 CEST

I hope the film is nice – however – rate me if you want.

Versionshelf just got better.

Thu, 19 Jul 2007 15:21:18 CEST

We just updated our subversion hosting service Versionshelf with the following cool new features:

much improved optics of the svn browser
intra line diffing in the svn browser (those of you who know FileMerge.app know how vaulble it can be to see exactly the 2 characters that changed)
international pricing in USD for non european customers.
all plans now have more features (more users and/or more repositries and/or more space)
the basic plan got cheaper

So check it out – it’s a 30 day free trial. And feedback is, as always, very welcome!

SubEthaEdit 2.6.4 is out

Wed, 11 Jul 2007 18:55:28 CEST

Mainly a bugfix release. But I especially like the triple click selection for strings and variables. Check it out! SubEthaEdit 2.6.4 – Full release notes – here an excerpt of the more interesting ones:

Additions/New Features

Added the AppleScript command "show" to the document. "show" will show the frontmost window / tab of a document.
Added the system wide show animation for selecting found text on Leopard
Triple click now selects up to style boundaries. E.g. strings, variables.
Save dialog now preselects filename without extension

Changes

When removing a split now, instead of the cursor position of the upper text view, the cursor position of the active text view is taken
Changed "Check Syntax" in all modes. It now should work with all encodings and with tabs.
Improved the syntax highlighting in Perl Mode
Improved Objective-C Mode to include support for Objective-C 2.0
Improved a behaviour where new windows would open across screens, now new windows open on the screen of the topmost window.
The report a bug menu item in the help menu now directly selects SubEthaEdit and the correct version in our bugtracker.

iPhone

Wed, 11 Jul 2007 00:41:01 CEST

Thanks to the imported iPhone of Alexander Olma of iPhoneBlog.de today I got my hands on an iPhone. And thanks to the nice hackers that provided a way of activating it without buying an AT&T contract. Of course we did talk about it at our german podcast, Bitsundso

My impressions:

everthing is way smaller as seen on the screenshots. I knew that the iPhone had a high resolution display, but everything was quite a bit smaller as I expected
the browser works great, however, selecting and deleting in large text areas really is missing. But I could post a nice bugreport at codingmonkeys.de/bugs and all the nice javascripty features worked fine
it really is super responsive. Especially the speed of flicking through the images on the phone was astounding. A big big contrast to every smartphone i used before.
the keyboard works great. I was able to instantly write long english notes quickly in the notes application with two thumbs. And even unknown words are easy to type if you do the push, move until the correct key is shown lift your finger then trick
I was surprised that the youtube app wouldn’t work without a sim card, obviously it looks for EDGE first which isn’t even an option without a sim card
it streams podcast nicely when going through the feeds with enclosure – however – no chapter marks or images for enhanced podcasts.
the calculater app is missing a possiblity to export the calculated number – again because of the lack of a copy and paste mechanism
the sms app misses a character count, so no way of knowing if you are over the 160 characters per sms. And because of the missing sim card I couldn’t test if it justs sends more than one message or merges longer messages as the nokia does.
although everything is quite small, all the buttons and even links in webpages are easy to hit, and hard to miss.
the mail app seems to miss a sent folder for pop accounts, i hope it has one for imap accounts
even when sending mails while in another app, the familiar Mac OS X mail sent sounds confirms that the mail was sent later on
great detail has been put in the way navigating works. Especially the way it animates when coming to boundaries when scrolling or zooming.

All in all a very nice device, and nothing short of what apple announced. I’m really looking forward to the european launch and I think I will be an instant buyer.

iPhone crash log information

Sat, 30 Jun 2007 22:17:17 CEST

Apparently when an app on the iPhone crashes, the phone notices after 30 seconds, writes out a crashlog and continues to work. On next connect with iTunes it wants to post the crashlog.

The contents of these crashlogs are interesting, as Frasier Spears already noted and found out.

So far I have found 2 crashlogs on the net. One at Daring Fireball and one at Martin Gordon – I think there will be more soon.

Here are the “Headers” of the two logs found up to now:

Process:         MobileMail [67]
Path:            /Applications/MobileMail.app/MobileMail
Version:         N/A (N/A)
Code Type:       0000000C (Native)
Effective UID:   0
Parent Process:  SpringBoard [15]

Date/Time:       2007-06-29 20:42:06.479 -0400
OS Version:      OS X 1.0 (1A543a)
Report Version:  6

Exception Type:  00000020
Exception Codes: 0x8badf00d
Crashed Thread:  Unknown

.hr

Process:         Preferences [73]
Path:            /Applications/Preferences.app/Preferences
Version:         N/A (N/A)
Code Type:       0000000C (Native)
Effective UID:   0
Parent Process:  SpringBoard [15]

Date/Time:       2007-06-29 20:27:24.929 -0400
OS Version:      OS X 1.0 (1A543a)
Report Version:  6

Exception Type:  00000020
Exception Codes: 0x8badf00d
Crashed Thread:  Unknown

Both crash logs look almost identical, both don’t know which thread crashed and both have the funny exception code of “badfood”.

Let us compare this to a standard Mac OS X Tiger crash log:

Host Name:      dom
Date/Time:      2007-02-08 15:48:36.933 +0100
OS Version:     10.4.8 (Build 8N1037)
Report Version: 4

Command: SubEthaEdit
Path:    /Appications/SubEthaEdit.app/Contents/MacOS/SubEthaEdit
Parent:  WindowServer [96]

Version: 2.6 (2043M)

PID:    414
Thread: 0

Exception:  EXC_BAD_ACCESS (0x0001)
Codes:      KERN_INVALID_ADDRESS (0x0001) at 0x73617283

So we can see there is no explicit PID mentioned on the iPhone, instead it is displayed in brackets behind the Process information (which seams consistent with the way it always was displayed for the parent process. The Report Version has bumped from 4 to 6. Also the OS Version scheme is different, although similar. It is not 10.something but OS X 1.0 – the running build number (1A543a) seems to be a similar scheme than the build number on tiger (Build 8N1037). The Parent Process for the Apps is the “SpringBoard” which seems to replace the “WindowServer” on Tiger. The Version of the current App seems N/A always – the Path to the binary seems to have lost a “Contents” which was quite superfluous anyway and is a old relic from next times that I never understood. Maybe they thought they would put metadata in the space between the .app folder and the Contents Folder… The “Effective UID” field is new, but filled with 0 doesn’t tell us much. Also new is the “Code Type” field – in which (native) would suggest that there may be other values which are not native…

Since we have no symbols I skip the thread part. After the thread part however, interestingly enough, there is no thread state in the logs, so no deducing of the CPU because of the displayed contents of the registers. The only text we get is:

Unknown thread crashed with unknown flavor: 4, state_count: 1

Moving on – The binary images list has a different format on the iPhone:

0x301c7000 - 0x301d4fff  AddressBook UUID (7D1B2DA0EC1D4A1382D5FE0B91851ACA) /System/Library/Frameworks/AddressBook.framework/AddressBook

against

0x94726000 - 0x9486cfff com.apple.AddressBook.framework 4.0.4 (485.1) /System/Library/Frameworks/AddressBook.framework/Versions/A/AddressBook

So instead of version numbers we now have UUIDs, the reverse DNS name syntax seems to have gone, and the complicated linked structure of the framework bundles also seems to have gone.

I want to close with the list of library/framework names which don’t appear on Tiger:

AddressBookUI UUID (0E6FBEAA0D9C411FB1F0794F35D2CFBA) /System/Library/Frameworks/AddressBookUI.framework/AddressBookUI
AppSupport UUID (4D84C1C39C7C4DB9981FEBF8E48FF450) /System/Library/Frameworks/AppSupport.framework/AppSupport
BluetoothManager UUID (E00ECB53D4B5426A818DA8D77CD1B09C) /System/Library/Frameworks/BluetoothManager.framework/BluetoothManager
Calendar UUID (5D3EB213B2A94C4C97C194283493122C) /System/Library/Frameworks/Calendar.framework/Calendar
Celestial UUID (15045615F83249D49720253EBDF9132A) /System/Library/Frameworks/Celestial.framework/Celestial
CoreSurface UUID (7280076DC074497CB145741A15FCE472) /System/Library/Frameworks/CoreSurface.framework/CoreSurface
CoreTelephony UUID (7B909435DABA409098995CC24EE26587) /System/Library/Frameworks/CoreTelephony.framework/CoreTelephony
GraphicsServices UUID (1C4876C189F34562ACF6B7D44770FF97) /System/Library/Frameworks/GraphicsServices.framework/GraphicsServices
IAP UUID (C298EEA4A51040D1959591C364206B54) /System/Library/Frameworks/IAP.framework/IAP
IOMobileFramebuffer UUID (AE80733EB6514BB88A6287971C7E09C9) /System/Library/Frameworks/IOMobileFramebuffer.framework/IOMobileFramebuffer
ITSync UUID (B02A31E7730A46BCA6B336F097D542C4) /System/Library/Frameworks/ITSync.framework/ITSync
LayerKit UUID (B9468BD4F4F041C3ADA50B830305E7A2) /System/Library/Frameworks/LayerKit.framework/LayerKit
MBX2D UUID (1583F2C2C78B4058BD8C1DF338738C05) /System/Library/Frameworks/MBX2D.framework/MBX2D
MBXConnect UUID (B5E73A95A84E4706A922D56674D4809D) /System/Library/Frameworks/MBXConnect.framework/MBXConnect
MeCCA UUID (5031D4A8F03B4605A6233D351EF790E0) /System/Library/Frameworks/MeCCA.framework/MeCCA
MessageUI UUID (ED6F6EA00B1346BEB98F00733C936922) /System/Library/Frameworks/MessageUI.framework/MessageUI
MobileBluetooth UUID (5F13101DF17B442D8DE5CD0BDB50D7AB) /System/Library/Frameworks/MobileBluetooth.framework/MobileBluetooth
MobileMail UUID (8859A31CE4874DF1AE2A9A92B1B717A5) /Applications/MobileMail.app/MobileMail
MobileMailSettings UUID (C7B98C13676647FA9A1A6D9DE36897E2) /System/Library/PreferenceBundles/MobileMailSettings.bundle/MobileMailSettings
MobileMusicPlayer UUID (C3D3AC1CC6384FDDB34D1D2D0A7E90F0) /System/Library/Frameworks/MobileMusicPlayer.framework/MobileMusicPlayer
MusicLibrary UUID (B9815425F5D742CD9B9FFF948A4E590D) /System/Library/Frameworks/MusicLibrary.framework/MusicLibrary
OpenGLES UUID (3405A88B732E4DF3A127E182483D9E69) /System/Library/Frameworks/OpenGLES.framework/OpenGLES
Preferences UUID (26FA104E4D88438285B988DDADDE39D6) /System/Library/Frameworks/Preferences.framework/Preferences
Preferences UUID (8E1045DED5F040138101BCE1C4C06AEC) /Applications/Preferences.app/Preferences
TelephonyUI UUID (15006830A8B94F7590AC10CBB664D0E7) /System/Library/Frameworks/TelephonyUI.framework/TelephonyUI
UIKit UUID (8A129F9979114365B637E144A3C67868) /System/Library/Frameworks/UIKit.framework/UIKit
URLify UUID (54B6437BCD3A452F9CC7493CB9E010FF) /System/Library/Frameworks/URLify.framework/URLify
liblockdown.dylib UUID (30A0EB0C05724FA487617B860E360B88) /usr/lib/liblockdown.dylib

That’s all for now :)

Update: Engadget identifies the processor used as an 620MHz ARM1176JZF which seems to contain the ARM Jazelle for embedded java execution. So the Code Type counterpart might be Java Byte Code.

Apple Introduces: Fixed width Pixel Fonts!

Mon, 16 Apr 2007 01:59:11 CEST

What the hell is happening in the Apple Pro App department? While Color might be a really great tool – why the heck did they use Amiga style fixed width pixel fonts all around the interface? Are we back to the days where Pro Tools have to look ugly, because they are Pro? Or did the Pro guys just not listen to Steve Jobs Stanford Commencement Speech of 2005 – just wondering…

Another German Thingy...

Tue, 6 Mar 2007 16:31:31 CET

My Thawte email certificate expired in January – so I went for a renewal. There I found out that my certificate is not the only thing that would need an update…

Music DRM be gone!!!

Tue, 6 Feb 2007 22:20:12 CET

Steve Jobs has given his opinion on DRM on the Apple Website :

The third alternative is to abolish DRMs entirely. Imagine a world where every online store sells DRM-free music encoded in open licensable formats. In such a world, any player can play music purchased from any store, and any store can sell music which is playable on all players. This is clearly the best alternative for consumers, and Apple would embrace it in a heartbeat.

Please, Music Industry, make the world a better place and let my purchased music be DRM free!

Amazon UnSpun: Best Text Editors

Fri, 26 Jan 2007 12:24:36 CET

Have a look and give your rating: Best Text Editors

At the moment of writing Textmate is Number 1 and SubEthaEdit is on Number 7 which is the second place in Mac Specific Editors.

Wii - some afterthoughts

Mon, 25 Dec 2006 15:36:19 CET

So I got my wii on the first day – here are some random thoughts about it:

Wii Play: Okay, what do you expect from a game that actually is thrown in with a controller. But seriously: most of the games would just need a little improvement to be really cool.

Rayman Raving Rabbits: This one I really like. It really shows what Wii Play could have been if they hat put more finish on the mini games.

Internet Channel: looks promising. What’s bad though is that it doesn’t support Bonjour. WHY? just so I have to enter 10.0.1.10 instead of brainy.local? Really Wii, just add Bonjour and you become my media center. I feel the urge to find a remote iTunes flash app, and a nice foto browser for iPhoto that just works web based and I got most of my media on TV – jay.

Zelda: Really nice. But more an evolution than an revolution. And Zelda really shows that more graphic power wouldn’t have hurt. It kind of feels like the blurry graphics of the N64 Zelda that always kept me from playing it. However I enjoy it a lot and forgive the wishi washiness of the graphics. I’m really looking forward to metroid on.

The Controller: What I really don’t like is, that if you have a game that uses the nunchuck, both of your hands are tied to the controller. you can’t just get a bite or a drink because either you can’t steer or you can’t move. hmmm. And I really like my junk food and drink when I play games.

Wii Sports: Tennis is kind of way to blurry to control. I really had higher hopes for it. But Baseball and especially golf are great.

Mii Channel: nice gimmick but what gives? I’d really like more support and use for them.

Wii Console Codes: It’s really a shame that both ends have to enter their respective wii codes to be able to send each other stuff. A simple invite and deny variant would have been way better. And I’m really hoping for great online gaming experiences. I hope spore goes wild on wii. Promising but far from impressive at the moment.

IE7 installer is a good parent...

Thu, 19 Oct 2006 14:41:04 CEST

...and asks you if you have been a bad boy before you get the treat:

MP4 video to crash Quicktime and therefore Safari with it...

Wed, 11 Oct 2006 15:30:58 CEST

The German popular video-blog Ehrensenf regularily posts MP4 videos in Quicktime that crash on Intel Macs. I filed a bug under rdar://4578460 on 08-Jun-2006 04:59 PM and it is still there. Embarrassing Apple. Really. Why am I writing this now? I got my Intel iMac because my DreckBook Pro needs a Major repair – don’t ask – and I wanted to watch this Ehrensenf Episode while copying stuff. And Booom – there goes the crash. I hope I can post some nice Apple things soon. Maybe when my DreckBook Pro comes back and is fully healed… for now I’ll watch Ehrensenf using Flip4Mac and WMV. What a shame.

And for the interested: the relevant part of the crash report:

Thread 11 Crashed:
0   ...le.QuickTimeMPEG4.component     0x98cd45c1 Reconstruct_8x8(unsigned char*, int, unsigned char*, int, short*, unsigned char, unsigned char, unsigned char, int, unsigned char const*) + 261
1   ...le.QuickTimeMPEG4.component     0x98cfe288 DecodeMBInter(int, int, short*, short*, unsigned char, macroblock_stuff*, INSTANCE_DECO*) + 1206
2   ...le.QuickTimeMPEG4.component     0x98cffbdc ReadACombinedVideoPacketSimple(long*, frame*, frame*, INSTANCE_DECO*) + 3052
3   ...le.QuickTimeMPEG4.component     0x98d01cae MPEG4DMPR_Decompress(WorkInfo*, INSTANCE_DECO*) + 2942
4   ...le.QuickTimeMPEG4.component     0x98cf5710 MPEG4_DDrawBand + 50
5   ...ple.CoreServices.CarbonCore     0x90ccbd57 CallComponentFunctionCommon + 513
6   ...le.QuickTimeMPEG4.component     0x98cf5106 MPEG4_DComponentDispatch + 176
7   ...ple.CoreServices.CarbonCore     0x90ccba3c CallComponentDispatch + 34
8   ...ickTimeComponents.component     0x98754639 ImageCodecDrawBand + 43
9   ...ickTimeComponents.component     0x97edc9dd BaseCodec_DecompressWorkFunction + 1006
10  ...ple.CoreServices.CarbonCore     0x90cd7bf8 InvokeComponentMPWorkFunctionUPP + 22
11  ...ickTimeComponents.component     0x97edaa83 BaseCodec_DecompressCallBack_Common + 1014
12  com.apple.QuickTime                0x9436482d ExecuteCallBack_priv + 125
13  ...ple.CoreServices.CarbonCore     0x90cdec97 InvokeDeferredTaskUPP + 15
14  ...ple.CoreServices.CarbonCore     0x90ce7bec DeferredTaskThread + 120
15  libSystem.B.dylib                  0x90023d87 _pthread_body + 84

I've had it with iChat video chat...

Fri, 29 Sep 2006 12:36:32 CEST

Every frickin’ fraggin’ time I want to use or demonstrate iChat Video chatting it doesn’t work. Yay for the build in iSight, big nay for the network coders of iChat and Airport Extreme. Yes I have an Airport Extreme Base Station and a Standard DSL line on my side.

That has to be enough to be able to do a video or audio chat!

But obviously it isn’t. I have all Apple Hardware here, one might think that this should make sure that a connection can be made. You can say what you want about Skype, but they got this one right. Skype always just works. iChat doesn’t even work when you have all Apple Hardware. Shame on you Apple! And all you did to improve the situation is provide a “Send to Apple” button if something doesn’t work. I sent and sent and sent. Now it’s enough, I’ve given up. Apple – when did you change from being the company where stuff just works?

update: An excerpt from the Using iChat AV with a firewall or NAT router – I highlighted the great statement. The Innovative feature is calles: Yourself did not respond, which makes perfectly clear who is to blame.

Network Address Translation (NAT) Some Internet service providers (ISPs) and home networking routers use a technology called network address translation (NAT) to share an Internet connection. Though this often interferes with video and audio connections in other conferencing applications, iChat AV uses an innovative approach to establish a direct audio and video connection even on networks that feature NAT. In fact, iChat AV works fine with many popular household routers in their out-of-box configurations. You can see a list of these routers in technical document 93333, “iChat AV: Compatible Network Routers”.

Applying Behaviour without Behaviour.js

Thu, 14 Sep 2006 19:01:34 CEST

In June 2005 Ben Nolan made my web-development live more joyful by releasing behaviour.js – a little javascript library that helps you keep your HTML clean. With behaviour.js you add all the javascript behaviour and magic after the page is loaded via CSS selectors.

So, instead of adding the onclick=”something()” attribute directly to the HTML you can put clean images into your HTML with classes like “removehandle” and apply this behaviour:

var often = {

    '.removehandle' : function(el) {

        el.onclick = function() {

            Element.remove(this.parentNode.parentNode)

            return false;

        }

    },

    '.optionlist_removeitem' : function (el) {

        el.onclick = function() {

            Element.remove(this.parentNode)

            return false;

        }

    }

}

Very nice. But what is even nicer is, that the current prototype version 1.5.0_rc0 features the $$() function which can do exactly the same in combination with the each() method. Interstingly enough, the easiest way to get your hands on this prototype version is the rails repository .

function applyRemoveBehaviour() = {

    $$('.removehandle').each(function(el) {

        el.onclick = function() {

            Element.remove(this.parentNode.parentNode)

            return false

        }

    })

    $$('.optionlist_removeitem').each(function (el) {

        el.onclick = function() {

            Element.remove(this.parentNode)

            return false

        }

    })

}

Versionshelf has landed

Thu, 24 Aug 2006 00:31:55 CEST

I’m proud to announce that Verionshelf is up and running today. Have a look, try it out. We offer a free educational plan, Secure Socket Layer throughout all plans and a 30-Day free trial period. So there is nothing to loose, but so much to gain. Once you version you never go back…

StopGoogle.com

Sun, 20 Aug 2006 19:20:05 CEST

Google is now actively warning about “malicious” pages. As you can see at stopbadware.org Google doesn’t just place a “bad” button next to your search result, but it shows you a page that looks a lot like a 404 to tell you to not go there, and if you really want to you can find the link further down.

Call me paranoid, but I think that is evil. Just imagine the damage caused by a wrong entry?

And very interesting, stopbadware actively reports that www.ThemeXP.org is evil but if you google for it, ThemeXP.org is the very prominent first hit which doesn’t get the nag screen. So just started and already some disconnect between google and stopbadware – what will stop anyone from making arbitrary decisions there?

You just gotta love my Mac OS X Issues...

Fri, 11 Aug 2006 22:50:10 CEST

Guess what, today my vlc crashed again on my MacBook Pro (which it frequently does when switching from and to fullscreen, or/and adding and removing secondary displays). But this time the crashing crashed:

Date/Time:      2006-08-11 22:39:58.661 +0200
OS Version:     10.4.7 (Build 8J2135a)
Report Version: 4

Command: Exited process
Path:    Exited process
Parent:  kernel_task [0]

Version: ??? (???)

PID:    2144
Thread: Unknown

Exception:  EXC_BAD_ACCESS (0x0001)
Codes:      KERN_INVALID_ADDRESS (0x0001) at 0x1887f000
(null)
Unknown thread crashed with i386 Thread State:
eax: 0x1887f000    ebx: 0x1c3036d2 ecx:0x03406ae4 edx: 0x1887f008
edi: 0x1d604200    esi: 0x034067bc ebp:0xb04fd578 esp: 0xb04fd560
 ss: 0x0000002f    efl: 0x00010206 eip:0x1c3b37d2  cs: 0x00000027
 ds: 0x0000002f     es: 0x0000002f  fs:0x00000000  gs: 0x00000037

Binary Images Description:

Model: MacBookPro1,1, BootROM MBP11.0055.B02, 2 processors, Intel Core Duo, 2 GHz, 2 GB
Graphics: ATI Radeon X1600, ATY,RadeonX1600, PCIe, 256 MB
Memory Module: DIMM0/BANK 0, 1 GB, DDR2 SDRAM, 667 MHz
Memory Module: DIMM1/BANK 1, 1 GB, DDR2 SDRAM, 667 MHz
AirPort: spairport_wireless_card_type_airport_extreme (0x168C, 0x86), 0.1.24
Bluetooth: Version 1.7.5f10, 2 service, 1 devices, 1 incoming serial ports
Network Service: AirPort, AirPort, en1
Serial ATA Device: TOSHIBA MK1234GSX, 111.79 GB
Parallel ATA Device: MATSHITADVD-R   UJ-857
USB Device: Built-in iSight, Micron, Up to 480 Mb/sec, 500 mA
USB Device: IR Receiver, Apple Computer, Inc., Up to 12 Mb/sec, 500 mA
USB Device: Apple Optical USB Mouse, Mitsumi Electric, Up to 1.5 Mb/sec, 500 mA
USB Device: Apple Internal Keyboard / Trackpad, Apple Computer, Up to 12 Mb/sec, 500 mA
USB Device: Bluetooth HCI, Up to 12 Mb/sec, 500 mA

WWDC 2006 Keynote

Tue, 8 Aug 2006 10:02:45 CEST

Hmmm… am I the only one that is a little bit underwhelmed by this Keynote? Mac Pro and Xserves, alright. They are at a good price point too. That they are “only” 2 times faster than the current fastest G5s kind of is a good testament to the G5. It seems it really was a nice Chip after all. Just too big and too hot.

What have we got. Well, notes in Mail ? WTF? Sounds like a feature Steve wanted. It doesn’t make much sense to me. A global To-Do list “Server” on the other hand sounds great, but putting the to-do lists mainly in mail on the other hand doesn’t really. But I love how the global Address Book turned out, so let’s hope they did it equally well. But Stationery in Mail? Yes my mother will love that (if it looks great in outlook on the receiving side and doesn’t get caught in the standard spam filters), but is that innovation? I don’t get the audience clapping on that.

Time Machine . I have very mixed thoughts on that. Could be great could be horrible. I hopes it comes with a successor to HFS+. And the UI? What’s up with that? I imagine a dual screen setup, and then an itsy bitsy tiny wheeny Address Book window in the middle. That just doesn’t work right. They wanted to show Core Animation , alright, but please please change that UI.

iChat Theater as well as screen sharing is IMHO the real deal. The first one makes video chatting so much more useful, the latter is ideal for support, which I need to give regularily to my mother and friends I brought to the mac. Real great. I hope it works faster than VNC and Apple Remote Desktop today.

And all that mocking about Redmond: Okay, Microsoft copies stuff. They do. But hey, now Apple postponed Leopard too. At least until Spring. I mean okay, we had a transition to Intel in between, that was much work. So be it. But Mac OS X is far from golden. There are so many things not working right, or as good as they could, please focus on that. Finder, Trashes, SMB-Support, Window Handling, OpenGL Speed, the Dock (when will it finally springload?), iChat AV connectivity, Laptop Heat, and and and. So stop mocking and continue working I say. I hope some of the Top Secret no-shows are really great, and the reason for not showing is not that they are just much too far from ready.

So what I’m looking forward to? iChat, Xcode 3.0, Objective-C 2.0, finally a significant update for Interface builder and of course, the top secret stuff. But what I’m really sad about is: With that timeframe, the next WWDC won’t be very interesting.

IB Wisdom: who knew?

Wed, 5 Jul 2006 18:37:07 CEST

Here I am, 3 years developing cocoa applications and working with Interface builder. And always I hated the fact that I had to resize all the interface elements when I wanted a different default window size. Little did I know that you actually can resize a window in IB with all the subviews autoresizing. Did you know this? And if, why didn’t you tell me?

Just hold ctrl while resizing. That does the trick. Haven’t that found documented anywhere, though.

CocoaDevHouse Munich: Results

Sun, 2 Jul 2006 14:14:46 CEST

It has been already a week now since CocoaDevHouse Munich happened. It was a great event – big thanks to Boinx for hosting it.

I worked together with Achim of Boinx on a Quartz-Composer only variant of Tetris. It was great fun, and although we had the game working on Saturday I spent some extra hours on Sunday to polish it so it can be released. But somehow I forgot to put it only somewhere ;). So here you go, the fruit of our labor: Compositris

We have been inspired by the AsteroidsQC that can be found on quartzcompositions.com. A great resource for nice compositions.

Be aware that Compositris is very demanding, best played on an intel Mac with built-in iSight. Achim was keen on doing much more visual stuff (neatly remove lines, and I already saw a great start game animation) and I’ll hope we get his enhanced version soon ;).

On another sidenote: I recently started the company Shelf Cloud which will soon offer reasonable priced Subversion hosting at Versionshelf.com – and we were happy to provide a repository for CocoaDevHouse Munich. If one or all of the other Dev Houses is interested we’ll add their attendees to the list of users for the repository we created or give an organisator admin access. I think it would be great if work on the started projects would continue between the dev houses, and versioning is a great tool for that.

Finally: I get to see .kkrieger

Thu, 15 Jun 2006 09:33:35 CEST

I heard about .kkrieger some time ago, but now on my MacBook Pro with Bootcamp I can actually play and see it. It’s amazing. All in 96k. That’s inspiring.

Big Fire in my Hometown

Wed, 14 Jun 2006 22:10:52 CEST

Time to use my .mac bandwidth. Two days ago, there was a big fire in my hometown, I captured the beginning using my Nokia N70 and put it on .mac. Really spectacular and dangerous. More pictures will be available on the local firefighters homepage. No one was harmed. The cause of the fire is yet unknown.

Update: The article on the firefighters page has been posted, and the video has been put on mytube for use at the Feuerwehr Weblog.

ZotCast on SubEthaEdit

Tue, 25 Apr 2006 13:51:40 CEST

Pom Pom Pom Pom has a nice advertisement and accolade of SubEthaEdit in his new ZotCast – and all in 5 different languages :).

Wocoa Frankfurt 2006

Mon, 10 Apr 2006 11:53:46 CEST

As mentioned on the TheCodingMonkeys Company Blog we’ve been to the Wocoa in Frankfurt. Although it was quite disorganized this time, we had some nice talks and I still would recommend the Event to anyone into cocoa or webobjects in germany / europe. Always a great place for socialising with fellow enthusiasts / programmers working in the field.

Mine and mbo’s photos can be found on flickr via the tag wocoa – I encourage the other attendees to also post their pictures there. And also to use OpenBC to add everyone. After all its all about networking :).

And I was happy to hear about Big Nerd Ranch Europe is holding a Ruby on Rails Bootcamp in Frankfurt. I encourage anyone that is interested in Ruby on Rails and has some training budget to keep an eye for their high quality classes.

All things monkey...

Mon, 20 Mar 2006 15:29:31 CET

...have now an own place: TheCodingMonkeys’ company blog. – have a visit and read about SubEthaEdit and our little Company.

Which sci-fi crew would I best fit in?

Sat, 18 Mar 2006 12:47:41 CET

You scored as Deep Space Nine (Star Trek). You have entered the dark side of the Star Trek universe. The paradise of Earth is far from you and you must survive despite having enemies on all fronts. But you wouldn't have it any other way because you thrive in conflict and will know what needs to be done to take care of those around you. Now if only the Founders would quit trying to take over the galaxy.

Your Ultimate Sci-Fi Profile II: which sci-fi crew would you best fit in? (pics)
created with QuizFarm.com

Deep Space Nine (Star Trek)

94%

Serenity (Firefly)

81%

Moya (Farscape)

81%

Babylon 5 (Babylon 5)

75%

Millennium Falcon (Star Wars)

75%

SG-1 (Stargate)

75%

FBI's X-Files Division (The X-Files)

69%

Nebuchadnezzar (The Matrix)

69%

Galactica (Battlestar: Galactica)

63%

Andromeda Ascendant (Andromeda)

50%

Bebop (Cowboy Bebop)

50%

Enterprise D (Star Trek)

38%

On getting the university degree

Thu, 16 Mar 2006 02:25:50 CET

In mid October 2005 I finally finished my university degree and I’m now able to call myself Diplom-Informatiker Univ. It took me about 6 years – 1 1/2 years longer than the regular time of 4 1/2 years. Why did I take longer – mostly because of 2 things: For one I worked a year without leave in an Web-Company (the .com boom payed so well), and of course there was the time invested in the creation of SubEthaEdit . I studied at the Technische Universität Munich which is considered one of the best in Germany. Now that some month have passed I think I’m able to do some retrospective. Things I’ve learned and want to pass on to all the new computer science students out there:

On learning how to be a good programmer
You simply don’t. There aren’t even lectures on programming or on specific programming languages. And the stuff you do in lectures and so called practical courses do not take you much further than any tutorial in any language does. In fact it is easily possible to get your degree without having to produce a single line of production quality code. If you want to learn coding you have to do it excessively in your spare time or get a part time job early on. And I think this is a major flaw in the german university education. Yes, abstraction and theory are important. But IMHO you need to have a good grasp of the basic skills in any profession to really understand the advanced concepts and theory.
Most lectures are bad or mediocre at best
I actually went to many lectures that were on topics of interest to me – and dropped out because they just made me tired and left me wanting for a good book. Of course there are the few exceptional ones, and I urge you to find them. To find the professors that are really enthusiastic about their subjects. This is something I’d also recommend for the final stages of school: choose by the teacher, not by the subject. A great teacher can make any subject interesting and valuable. The contrary holds true too: a bad teacher can totally spoil all your interest and enthusiasm about a subject. There is the argument that students are getting less and less interested, and that’s also true to some extend. But the average lecture quality really has to improve.

Having said that, I want to mention the people at the TU Munich that really had an impact on me:

Dr. Peter Vachenauer who held the Higher Mathematics lectures in our first year. Not only is he and educator par excellence but he also has just a phenomenal breadth of mathematical wisdom. He was able to show me all the interconnectiveness between the mathematical disciplines and the essence of some great mathematical concepts. He put an enormous efford into his teaching and it just showed. And he takes pride in his work.
Prof. Dr. Christoph Zenger who showed me that there are professors out there, that really cared that students get the best options for their educational career.
Prof. Donald Kossmann who really did the best lectures at the TU Munich. Sadly he only held three and then went on to other universities. I regret deeply that I missed his first lecture.
Prof. Bernd Brügge Ph.D. whose enthusiasm for the Apple Macintosh and his decision to have an open Mac Lab gave me the exposure to Mac OS X that I needed to find an successor to my Acorn Risc OS machines.
Dr. Rafael Kobylinski who gave me the opportunity to explore and learn Mac OS X technologies in an university context by doing many courses, lectures and projects.

And finally some things I really would like to see in computer science university education:

Getting Real
Let practical courses be real practical courses with companies and real projects – not extended tutorials. Show real world examples to go with the theory. If you talk about databases and indexing take the time and understanding of some real open source code out there. Little sandboxes of pseudo code are good for a basic understanding – but the reality differs enormously. When doing techniques like UML diagrams – show them on real world projects – not only praise their benefits, but also show their deficiencies. Teach things on the edge of current standards, not things already deprecated.
Teach basic skills
Teach that the choice of language does matter – that there are different programming languages out there that have their specific place for speicific tasks. Provide additional courses for the main programming languages and teach more than one. Teach about bug tracking and versioning and the basic tools needed. Recruit students for that – students that know because they we’re interested in the subject. Let them teach and get credit points for it.
Embrace Open Source
Teach that one man can make a difference. Teach how to read code. Teach how to contribute and fix bugs. Teach how to start an open source project. Teach how to take different open source projects and combine them into a new exciting one.
Teach passionate and from experience
Stop making the professors do lectures about subjects they obviously don’t care or know about. It shows and it doesn’t help. Good educators can only teach about subjects they are really into.
Take interest and value initiative
Inspire students to create software in their spare time, talk to them, lead them, take them to the point at which they become expoerts in a field, appreciate that.

mist blogs

Wed, 15 Mar 2006 17:18:54 CET

Fellow ex-tu-student and Xbox hacker – developer Michael ‘mist’ Steil has joined the ranks of bloggers together with Melissa Mears at pagetable.com – read and enjoy heavy intel assembly :)

Adobe is teh suck...

Wed, 15 Mar 2006 16:59:05 CET

In case you wondered if to install the flash security update on an intel based mac: don’t. Adobe doesn’t mention anything about being universal or not, but just installs on a MacBook Pro – after that you only have flash in rosetta. Once again: Big booo booo Adobe…

MacBook Pro - Sleeping / VLC crash

Wed, 1 Mar 2006 13:45:49 CET

Just two Things:

First, my MacBook Pro sometimes wakes from sleep very slowly – something I didn’t experience with my TiBook. My guess is that this is related to hibernation, although I don’t know why the machine would hibernate when the batteries are nearly full. We are talking about 10-15 seconds here to show me the display instead of the usual few.

Second, my use of the VLC nightly build of February 23rd came to an astounding halt, show the display in a state that I only know from turning of my GameBoy Advance (that slow fade to white). It happend after i hid VLC while being in the middle of a show – then some half our later I clicked on it. Pictures taken with my N70:

MacBook Pro - No Luck...

Thu, 23 Feb 2006 22:29:35 CET

My MacBook Pro arrived today – after all the sweat of moving my 90 GB from my old Titanium PowerBook G4 800. Did I mention I have no luck with apple machines? I think I did. Sad me – this time is no different:

My right speaker seems be broken – the left one is fine and good sounding – the right one distorted as if totally over powered or connected cross over. oh. that just might be the case. Has anyone the same issue? I bet – here is a thread in the support forum: Problem with built-inspeakers – sad sad thing. Nice remote for iTunes listening and movie watching, but awful sound.

Oh and a sidenote – when wanting to transfer my account via the “normal” way, instead of using ditto on the users home directory, this is what showed:

Yes, Mac OS X is thinking I got 9802631.3 GB on my TiBook…

P.S.: I also had my first kernel panic – yes – although maybe it was not wise to play world of warcraft to calibrate the power manager – when the battery hit zero, the machine paniced.

Thu Feb 23 23:52:12 2006
panic(cpu 1 caller 0x0019C9BE): Unresolved kernel trap (CPU 1, Type 14=page fault), registers:
CR0: 0x8001003b, CR2: 0x13c03d18, CR3: 0x00b26000, CR4: 0x000006a0
EAX: 0x0290ca00, EBX: 0x0290ca00, ECX: 0x0000000e, EDX: 0x23ef8038
ESP: 0x13c03d18, EBP: 0x13b3bf84, ESI: 0x13c03d18, EDI: 0x2a7bc0b8
EFL: 0x00010206, EIP: 0x2a7bc0dd, CS:  0x00000008, DS:  0xd1860010

Backtrace, Format - Frame : Return Address (4 potential args on stack) 
0x13b3be24 : 0x128b5e (0x3bb060 0x13b3be48 0x131bbc 0x0) 
0x13b3be64 : 0x19c9be (0x3c04e8 0x1 0xe 0x3c02a0) 
0x13b3bf14 : 0x197a43 (0x13b3bf28 0x13b3bf84 0x2a7bc0dd 0x310048) 
0x13b3bf20 : 0x2a7bc0dd (0x310048 0x250010 0x10 0xd1860010) 
0x13b3bf84 : 0x13d683 (0x290ca00 0x13c03d18 0x134aec 0x13b3bfd8) 
0x13b3bfd4 : 0x197911 (0x0 0x0 0x3ae2d48 0x290ca00) Backtrace terminated-invalid frame pointer 0x0
      Kernel loadable modules in backtrace (with dependencies):
         com.apple.iokit.IOUSBHIDDriver(2.4.0)@0x2a7ba000
            dependency: com.apple.iokit.IOUSBFamily(2.4.0)@0x29c8d000
            dependency: com.apple.iokit.IOHIDFamily(1.4.6)@0x23e0b000

Kernel version:
Darwin Kernel Version 8.5.1: Mon Jan 30 21:07:08 PST 2006; root:xnu-792.8.36.obj~1/RELEASE_I386

SubEthaEdit 2.3 and BugShelf

Mon, 6 Feb 2006 17:19:38 CET

It’s out, we released SubEthaEdit 2.3 today. Check it out. It’s full of improved UI and over 35 fixes, changes and additions – we also changed our license from free for non commercial use to free trial only.

Which brings me to my next subject: BugTracking. You know we used Atlassian’s Jira for quite some time now and were very outspoken about it. We want to thank them again for providing us with their great bugtracker. But since we changed from initially free software to the full commercial software now, this time is over. And although we really like Jira over the time we developed some urges for some ajax goodness, had some own ideas and felt that Jira was a little over the top for us. So this is what we did: we did our own Bugtracker – which is trimmed for simplicity yet have as much features as we needed. First thing you will note is, that it isn’t totally public anymore. Much like Apple’s bugreport.apple.com you can only see the bugs you reported. But I stepped it up a bit in that you can also see the bugs you reported duplicates for. And of course you don’t need to register to report a bug. However, if you report a bug and provide an email address, you can register using the confirmation email. And we have tags ;-).

We migrated all our Jira data and accounts, so if you had an old account, you can use it again to check up on all your reported bugs. All the old bugs are there, we changed the HYR-XXX naming scheme to SEE-XXX while keeping the numbers.

So that’s it, no more java on our little server. Please tell us what you like / dislike about SubEthaEdit at our Feedback form and file bugs and feature requests at our BugShelf .

lighttpd, apache 2 and rails

Fri, 20 Jan 2006 11:55:07 CET

James Duncan Davidson has written about Reverse Proxying SSL Connections To Rails and identified one problem we ran over too. Actually I have a different solution for the simple case in which everything should be https. Just put this code in your environment.rb after the requirement of boot:

#https patch

class ActionController::AbstractRequest

  def protocol

    'https://'

  end

end

Moreover in such a setup make sure that you are running an apache 2.2, because there is a nasty bug in the mod_proxy with 2.0.55 that totally screws up data when used in conjunction with https. Bug 37145

Music: Linda Perry - In Flight

Wed, 11 Jan 2006 18:26:03 CET

Finally: New Music from the ex-front-woman of the 4 Non Blondes Linda Perry . Her voice is just incredible. I got my hands on Fill Me Up some time ago, and it has been one of my favorite songs ever since. Checkout the album, it is really great.

My new MacBook Pro...

Wed, 11 Jan 2006 09:29:50 CET

I just ordered mine. My now over three years old TiBook just has to be replaced. Although the new name made me think twice. However, for those of you in Europe thinking of buying one and wondering what “February” means: This is what the german apple store predicts (and I made my order the earliest point it was possible, 8.00 o’clock in the morning, hanging 25 minutes in the waiting loop. I had to make it via phone to use my ADC-Select Hardware discount)

Ihre Bestellung wird voraussichtlich am 15.02.2006 versendet.

And some thoughts on the keynote:

Why the hell did steve show safari as a speed benchmark after photoshop?
I would have never thought they replace the new iMac models this quickly.
Is the fast iMac now the fastes Mac? I surely bet so
Enjoy the speed bump folks, we won’t see another speed bump that gets even close to twice as fast any time soon.
Wood or marble 3D charts in iWork? I’m sorry, somebody lost it’s taste there…
I want my numbers… but at least now I can use Pages instead of word for my invoices (calculated values in tables), yay. After I pay another 79 EUR…

Update: Oopsi, closing the lid sends the current ones into nirvana, look at these comments – hopefully they fix all issues until they ship.

My #&!+9$!# iMac 20''

Sun, 18 Dec 2005 11:54:55 CET

I’m totally out of luck with my Macs. Every Mac I owned had its problems. First the iBook G3 (first white iBook), logic board failure. Then my powerbook: a broken RAM that wasn’t detectible via tests, but shuffled some bits on my harddisk, and every two to three month my machine was broken because of that. Then the Powerbook 12’’ for my mother. Had some sleeping problems – sometimes didn’t goto sleep, sometimes hung in an endless loop with full fan activity when put to sleep. And the iMac 20’’ for my father. I encountered the problem described in this knowledge base article . Ok, so repair was the way to go. Yes it was repaired and it was on warranty. But now, some month later, the same problem again? WTF? did they replace the parts with the same faulty but newer ones? I don’t know, and hope that the repair extension program still is in place and will make this repair cost free. I really hope this.

Boot Screen

Here we get stuck…

Single User Boot

Update: The first time I actually get an error code from the Apple Hardware Test: 2NVD/1/4: 2103 – probably the nvidia graphics card…

The barely readable results of the Apple Hardware Test

Update: the iMac got repaired quickly and on warranty. Hope this time in wil work a little bit longer than just a few month…

Apple Bug Report Someday: Application Activation fails in certain occasions

Mon, 12 Dec 2005 16:39:56 CET

Bug #4364207

Summary

If you change between applications, and do the activation click as a cancellation of a context menu you generated via a right click, then the application can’t be activated anymore and the menu bar does not react.

Steps to Reproduce:

open Textedit and a new window
open Finder
make sure the finder is active and the textedit window is visible
right click on the textedit window in the text area, see the standard context menu of a TextView
now cancel that menu by left clicking somewhere in the textedit window
see the textedit window raise, but the window is not activated, neither is the menu of textedit shown
now you can switch between windows of textedit and nothing works
if you press apple-b for bold, then the menu blinks and appears, but only certain items are enabled, and most of them don’t do any work

Expected Results:

After cancellation of the context menu I expect normal activation of this window and app

Actual Results:

The application gets into some kind of almost-active zombie mode, that can only be escaped via changing to another application and back

Notes:

This behaviour can be reproduced also with safari or any other app that produces a context menu on right-click.

Please please don’t fix this feature by first activating the app and then sending the right click. This would be a real bad behaviour too.

Rails Tidbits: Association Proxies

Sun, 11 Dec 2005 14:43:51 CET

I’m astounded how sure I can be about how things worked, and then be totally wrong. That was the case with Rails and association behaviour. After the latest upgrade, I thought something broke which in fact never worked the way I thought it works.

As I was sure about it and thought it was a newly introduced bug I quickly pested the rails folks with this Ticket 3178 bug-report. At least I provided my informations, but sorry for the big hassle. Now to some explanations why this snipped works this way:

>> comment = Comment.find :first
=> #<Comment:0x2735284 @attributes={"name"=>nil, "id"=>"1", "user_id"=>"1", "status_id"=>"1"}>
>> old_value = comment.status
=> #<Status:0x2730a40 @attributes={"name"=>"ok", "id"=>"1"}>
>> new_value = Status.find 2
=> #<Status:0x272a794 @attributes={"name"=>"not ok", "id"=>"2"}>
>> comment.status = new_value
=> #<Status:0x272a794 @attributes={"name"=>"not ok", "id"=>"2"}>
>> old_value
=> #<Status:0x272a794 @attributes={"name"=>"not ok", "id"=>"2"}>

As you see the old_value somehow magically sticks to the value of comment.status, not something you would expect. Especially because even the object ids are different. Why is this the case?

Acutally its rather simple, when you add belongs_to to your model, then accessor methods are created. These accessor methods have a little twist on them, that is the object actually returned is not a model object, but an AssociationProxy object (active_record/associations/association_proxy.rb)- this association proxy really acts as the targeted model object via mainly this piece of code:

alias_method :proxy_respond_to?, :respond_to?

def respond_to?(symbol, include_priv = false)

  proxy_respond_to?(symbol, include_priv) || (load_target && @target.respond_to?(symbol, include_priv))

end

This code actually redirects all of the methodes to the target of the proxy. Pretty neat. So while looking like an actual model object what you have is a proxy object. Moreover, the setter method added by the belongs_to method doesn’t replace that proxy, but merely updates its target. So, one side effect is the behaviour I thought was an bug. Now that I know, I can easily cope with it. However, I think it would be more appropriate if the Accessor method actually would return the target of the association, instead of the Proxy. On the other hand by this behaviour the Proxy can delay the way to the database.

=> #<Comment:0x2749bd0 @attributes={"name"=>nil, "id"=>"1", "user_id"=>"1", "status_id"=>"1"}>
>> old_value = comment.status.target
=> #<Status:0x27470d8 @attributes={"name"=>"ok", "id"=>"1"}>
>> new_value = Status.find 2
=> #<Status:0x27449c8 @attributes={"name"=>"not ok", "id"=>"2"}>
>> comment.status = new_value
=> #<Status:0x27449c8 @attributes={"name"=>"not ok", "id"=>"2"}>
>> old_value
=> #<Status:0x27470d8 @attributes={"name"=>"ok", "id"=>"1"}>

So bottom line: be aware that model objects returned by association accessors aren’t the real deal, but proxies, which update with the model.

Joel on Interns

Tue, 6 Dec 2005 09:46:32 CET

I watched my copy of Project Aardvark yesterday. BTW: really nice blog layout. When I first heard of that Joel Spolsky was making a documentary about an internship at FogCreek software, I instantly ordered my copy. Well, after I watched the trailer ;-).

After watching the film I have to say I really liked it. Although I would have loved to see more coding and coding problems and less fun-making. Nevertheless a very enjoyable film, and Paul Graham’s guest appearance really is a plus. The quote I enjoyed most was by Aaron Swartz :

You have this experience to this outside world, right. And like, you sat down and you’ve built software that seems useful. You feel like, well, you could be part of this outside world, you don’t have to be in this sort of fake world of school doing what: you know, some silly assignment that has no real purpose. You could build something that’s actually useful. You can go put it up to your website and people can really use ist. So I think thats what you really like. If you can build something real, why spend time doing stuff that’s fake?

So go ahead and buy it! I’d like to see more of this type of documentaries!

Ruby Observations: Hash#select vs. Hash#reject

Sun, 4 Dec 2005 16:47:01 CET

hash = {:key1 => 'value1',:key2 => 'value2'}

=> {:key1=>"value1", :key2=>"value2"}

hash.select{|k,v| [:key1].include? k}

=> [[:key1, "value1"]]

hash.reject{|k,v| ![:key1].include? k}

=> {:key1=>"value1"}

So the result of an select is an array of 2 element array, where the result of an reject is a hash. Useful but strange. I was already on my way to implement something like Hash#select_by_key because I saw the behaviour of Hash#select.

Apple Bug Report Friday: Safari: Canvas element.

Sat, 3 Dec 2005 21:24:49 CET

My first entry for the Apple Bug Report Friday, may seem to be a nuisance, but IMHO really symptomatic for Mac OS X.

Bug #4364207

Vector art of the canvas element gets rasterized on print

look at here

there is a perfectly drawn vector monkey. real nice. now you want to have that in pdf. Simply press print and preview? sadly no. then you get a pixel image, although the canvas element is filled using vector commands.

This is getting quite frustrating. There is Mac OS X with Quartz, so potentially everything is vector, and could be full quality. Vector graphics really is treated stepmotherly all over the place. All the technology is in place, please provide the tool-chain to actually get out as much vector art as possible, without all that rasterization. Much of the time it is easier to get decent vector art on my old Risc OS machine than on my Mac.

sidenote: please make the svg support right. so you can put svgs in img tags and as well print them as vector art.

Microsoft Office 12

Fri, 18 Nov 2005 07:30:32 CET

After my first look at the screenshots of Microsoft Office 12 I was disgusted and thought wtf? But after I watched the promo video and the other stuff on the net, I understood where they are heading. It may not be very good, but I think it will be a big improvement over the previous overloaded UI design. But what really hit me is, that Microsoft is at least trying to change some UI stuff fundamentally.

I mean look at Keynote and Pages. What do we have? Inspectors and Palettes and much too big toolbars. That doesn’t work. It’s annoying in Photoshop, it’s annoying in OmniGraffle and is annyoing in Pages and keynote. Ever tried to edit a keynote presentation on a 12’’ powerbook on 1024×768? No way you can adjust your window setting in a pleasant way to work. Inspectors are old, and while an quite ok idea, the screen clutter they create is really a pain in the as. I really would like to see Apple to innovate in that area. Microsoft did in its own way, by forcing the inspectors in the upper toolbar row and making this row very context sensitive. I think its a big improvement and a step in the right direction. It will be very interesting what the next Office for mac will look like.

New Website, new Logo, new Language

Thu, 10 Nov 2005 14:16:07 CET

Hi everybody! Hallo zusammen!

Our new website is up, featuring our new Logo! I hope you like it, comments are very welcome.
Seit Gestern ist unsere neue Website online. Feedback ist erwünscht.

The Logo was designed by Frank Röpke of the DesignerInNot and we are very happy with it.
Das Logo ist von Frank Röpke, einer der DesignerInNot und wir sind sehr zufrieden.

And we finally made our homepage and the SubEthaEdit page bilingual. As is this blog. As I know not many of my readers are bilingual, so I added 2 rss feeds – one for all german articles and one for all english articles. Articles like these with weird bilinguality will show up in both feeds.
Endlich haben wir auch deutsche web-seiten. Sowohl die Firmen- als auch die SubEthaEdit Seiten sind jetzt zweisprachig. Und da ich auch gerne mal deutsch oder englisch blogge, hab ich mein Blog jetzt um die Kategorien english und deutsch erweitert, zu denen ich auch eigene RSS feeds gemacht habe (ganz unten auf der Seite und rechts in der Seitenleiste). Zweisprachige Einträge wie dieser erscheinen in beiden feeds.

Back to blogging...

Thu, 3 Nov 2005 08:24:51 CET

Hi everybody!

I’ve been quite busy the last few month to get my university degree and therefore neglected my blog somewhat. This will change again now. With the degree in my pocket I will concentrate on TheCodingMonkeys and my other ventures. And on my way I will blog along regularly again. So stay tuned!

Pom Pom Pom Pom mit TheCodingMonkeys

Fri, 21 Oct 2005 01:08:25 CEST

Pom Pom Pom Pom ist ein witziger deutscher Mac Podcast. Und wir waren diesmal dabei. War ein Riesenspaß und ich hoffe ihr habt soviel Spaß beim anhören wie “Affen die coden” beim machen hatten. Wir haben das mit einem netten binauralen Micro auf einem Basketball gemounted aufgenommen, so dass mbo rechts ist, map links und ich in der mitte.

We have been on the german podcast Pom Pom Pom Pom , if you understand german, check it out. We did it with an binaural microphone mounted on a basketball, mbo is on the right, map on the left and I am in the middle.

New iPod...

Wed, 12 Oct 2005 21:07:34 CEST

Just to be the first to have said it: no more firewire for your iPoddy… so no new iPoddy for my old PowerBooky… (30GB via USB 1.0? don’t think so).

And no more remote control, which is equally bad for me in my car. So definitely no new iPod for me anytime soon.

Finally: Xara will come to the mac!

Wed, 12 Oct 2005 09:42:15 CEST

After years of begging the Xara people for making a Mac version they really will do it. More so, they will make it Open Source. For Linux too. I hope this isn’t a hoax. Finally a decent creative vector graphics package will come to the Mac. For me it is christmas already…