Planet Larry

Jürgen Geuter: Openness as a quality

Wed, 29 Sep 2010 21:29:53 +0000

A few days ago I wrote about the release of the open-source, distributed “clone” of Facebook called Diaspora in an article I headlined “Diaspora fail“. Others have taken it apart even in more detail than I have showing that there are more than just bugs but huge conceptual problems with diaspora but in this text I don’t really want to address those implementation problems at all. I’ll look at the “openness” as a feature.

People were looking forward to Diaspora as their savior, the one to battle the Big Bad Inevitable that is Facebook. Everybody could have diaspora on their own machine being in full control and the software would be open and plugins and addons and features would blossom all over the Internet, leaving Facebook in the dust.

Join me and look at recent history. A few years ago Twitter was the current darling of the scene and it was where Facebook is now: Everybody used it but it was closed, you couldn’t modify it or run it on your own and you had to give your connections and data to a company. But somebody (In this case Evan Prodromou) stepped up to the task and implemented a free clone called “Status.net” (it used to be called laconi.ca). Status.net is technically superior, it has groups, it is distributed, it can be used via XMPP and you can easily (well kinda easily but that’s always the case) slap it on your own server. But only few people use it, everybody is still on Twitter. Why? Because everybody is.

This sounds circular and it actually is: A community can be held hostage. If you have all the people, nobody wants to leave cause you don’t wanna be locked away from your friends or the interesting people. Twitter has the movie stars and whatnot, so no matter how open or technologically your thingy is, you will not beat the behemoth.

So maybe with Facebook-clones it will be different? Maybe Twitter was just a different case?

This is based on a misconception that I want to clear up: Openness does not always solve your problems, especially not if they are about privacy and social networks.

The problem with Facebook is not that it’s not open source. The problem is that it doesn’t help you keep stuff private but tries to get you to put more data out into the public that you probably want. The problem is that people accidentally upload a naked picture of themselves that is quickly mirrored before they can fix it. The problem is that people post 10 pictures of themselves being completely drunk and go to a job interview a week later. The problem is not the software and its lack of openness, the problem is the people using it and how the software treats them.

Even if Facebook was open source and somebody would fix the privacy setting mess, would that help? You’d still have to completely trust the guy who runs your OpenFacebook installation: If I host an OpenFacebook and you are on there and have everything private, I can still read all your stuff, copy your data, pictures, use your access token/key to access your friend’s data on other servers and copy those. All that would change is that instead of having one Big Bad, you’d have 100 or 1000 with a different scandal popping up with each and every different OpenFacebook installation.

As soon as somebody can access your data (we call those people “friends” or “contacts” in online lingo) you have to consider that data public. Somebody could hack their accounts or use his/her power over the machine their accounts are on to crawl your profile. The Openness of the platform does not matter, not at all.

With the success of many open source projects like Linux, Firefox and others the word “Open” has been imbued with a very positive aura. “OpenWhatever” is always better than just “Whatever”. But look around: Facebook’s OpenGraph sounds good, but apart from an API it’s not open at all!

“Open” has been turned into a marketing word (the process of labeling closed things “Open” has been called open-washing) but even for projects that might be truly open the openness does not always validate the project.

Diaspora and similar attempts are conceptually broken because instead of having to trust one specific entity you have to trust many other else. Diaspora fails not cause of the horrible mess of code they have and their lack of knowledge about basic web development techniques, it fails because it cannot win on a conceptual level.

Feel free to Flattr this post at flattr.com, if you like it.

Steve Dibb: planet larry migration

Wed, 29 Sep 2010 19:46:33 +0000

I wrote not too long ago that I was going to give up maintaining Planet Larry and look for a new owner. I got a lot of offers from a lot of people who were willing to step up and help take care of the project for me -- thanks guys!

Of all the submissions, one really stood out in my mind as someone who had some really great ideas for the site and was excited to take on the project -- Daniel Robbins. Daniel is the original founder of Gentoo, and currently maintains Funtoo. He and I have worked together in the past, and despite the controversy his name may bring being attached to the project in the eyes of some, he has assured me that he will run things objectively. I trust him, and believe there won't be any issues.

We're still working on the migration, which mostly means waiting for me to get some free time (crazy schedule), and part of that was announcing publicly that we have a new admin for the site.

Thanks, guys, for all your help in running the site. It's been good for me to work with so many users. I always enjoy getting another perspective on how Gentoo projects are used in the wild, and this has helped a small bit to share that view.

Jürgen Geuter: Transparency

Mon, 27 Sep 2010 20:53:22 +0000

When it comes to code many of you will know Linus’ Law:

“given enough eyeballs, all bugs are shallow.”

It basically states that you just need enough people to look at something to have them find what’s wrong with it and we’ve used that principle in free software development for many years now with (as I’d say) a lot of success.

In the last few days I’ve thought a lot about this idea, the idea of transparency. Now we shouldn’t make the mistake of mixing privacy and transparency up: Privacy deals with individuals, transparency is for organizations. So while we do want organizations to make their own decisions just like people, organizations do behave and should be treated differently.

With the Internet having established itself as cheap and omnipresent way of transmitting information, of reaching many many people (given you reach the right multiplicators) we’ve also seen many organizations and groups of people step up to the task and fight for transparency: The EFF, the CCC or projects like the now famous Wikileaks (among others).

They fight governments and cooperations who hide the “truth” from the public, who make deals against the public interest or funnel away public money into their own pockets and we all should be thankful for those organizations and their work. The world needs whistleblowers and people who dig in the darkness till they find what’s what.

But recently we’ve also seen the other side: Wikileaks’s completely secret organization and actions have lead to one person becoming the face and the boss of the project with the power to kick everybody out who might stand up against him. But wikileaks is far from the only “good” project that lacks the transparency that it so fiercely wants others to display. The CCC has a track record of being far less than stellar when it comes to being transparent about what’s going on (especially with the way the handled ticket sales [there were no tickets, then there were just for people that knew people, then there finally were some, it was a mess] and speaker slot assignment in the last few Chaos Communication Congresses [one of the public faces of the CCC said while pointing at the CfP in a podcast lately that not all slots were taken yet ... well since the deadline has not run out yet, shouldn't all slots be free still?][the review process is also far from being what we can call a "peer review" process or otherwise open and transparent]).

I started this text with a law or quote from a programming background, so let me add another one of those:

“Eating your own dog food“

Eating your own dog food means that a producer of something uses the product him- or herself: Microsoft uses Windows and MS Office on its machines and the bosses of Volkswagen are not supposed to drive around in Mercedes-Benz cars.

I think it’s time to start eating our own dog food. And I don’t just mean the organizations that I pointed at (and many other similar organizations) but each one of us: Why do we cheer for Wikileaks with every big leak (and publicity stunt) when they themselves do nothing to be transparent? Why do I still propose talks each year to the Chaos Communication Congress knowing that they will be rejected without any comment whatsoever? Why don’t we hold us and the organizations we cheer for at our/their word?

Transparency is the only way to build strong organizations. And you want strong organizations and not just strong people. People are easy to discredit (some even easy to corrupt) and too much power has never done anyone any good. We need good organizations and structures that make sure that power is equally distributed and that a structure has ways to make sure that every member behaves within the limitations of what the structure establishes. And the only way to make sure that people can check that nothing filthy is going on is transparency.

If you want to fight for transparency start with yourself and your own project. We need transparency everywhere but why should I trust one organization fighting for transparency if that organization itself isn’t?

Feel free to Flattr this post at flattr.com, if you like it.

Joachim Schiele: no more crashes after resuming from pm-suspend III

Mon, 27 Sep 2010 11:47:17 +0000

source hp.com

i’m still having several issues with this laptop, will it ever end? at least it’s not the nvidia proprietary driver again. so here are some details:

used kernel: 2.6.34-gentoo-r6

problem:

graphics causes problems with ‘sata – ahci’ mode
sound sometimes crashes the system (module load/unload)
wlan sometimes crashes the system (module load/unload)
acpi, not all keys report acpi events

graphics

used driver: nvidia-drivers-256.53 (nvidia proprietary)

so far most things i need are working, which includes:

2d acceleration (xvid extension using mplayer)
3d acceleration
console (alt+ctrl+f1) still works after pm-suspend’s resume cycle (not when no X was running prior pm-suspend)
resume will resume X without issues

untested:

external monitor setups as vga/hdmi
compositing: i still have compositing disabled as the system is much more responsive without it. still it would work well.

problems:

a new problem is the issue with the nvidia module and the sata ahci mode:

Sep 25 18:24:56 ebooK kernel: [ 2989.757123] ata1: exception Emask 0×10 SAct 0×0 SErr 0×10000 action 0xf

Sep 25 18:24:56 ebooK kernel: [ 2989.757134] ata1: SError: { PHYRdyChg }

Sep 25 18:24:56 ebooK kernel: [ 2989.757146] ata1: hard resetting link

Sep 25 18:24:57 ebooK kernel: [ 2990.480057] ata1: SATA link up 3.0 Gbps (SStatus 123 SControl 300)

Sep 25 18:24:57 ebooK kernel: [ 2990.482352] ata1.00: ACPI cmd ef/10:03:00:00:00:a0 (SET FEATURES) filtered out

Sep 25 18:24:57 ebooK kernel: [ 2990.485490] ata1.00: ACPI cmd ef/10:03:00:00:00:a0 (SET FEATURES) filtered out

Sep 25 18:24:57 ebooK kernel: [ 2990.485935] ata1.00: configured for UDMA/133

Sep 25 18:24:57 ebooK kernel: [ 2990.485945] ata1: EH complete

Sep 25 18:24:57 ebooK kernel: [ 2990.486235] sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn’t support DPO or FUA

[1] seems to confirm this problem. when i disable ahci (and use ide mode) in the bios this error message is gone. currently i don’t know how DPO or FUA or this whole error influences my stability. however: when my system crashes i have a very high chance of file loss, this means: all kde settings are deleted, chromium restores to factory defaults and similar problems.

sound

00:1b.0 Audio device: Intel Corporation 82801I (ICH9 Family) HD Audio Controller (rev 03)

the soundsystem crashes the system when snd_* modules are reloaded using ‘rmmod’ or ‘modprobe -r’. a trace can be found at [2].

this crash happens even if there was no pm-suspend cycle yet
it does not matter if sound was played (mp3 or film), it crashes anyway

-> using /etc/init.d/alsasound restart does NOT remove loaded modules and it does NOT shut down programs using the soundcard prior to removing the modules. but i’m using a script which does so.

i’ve written a mail to alsa-devel@alsa-project.org and i’m still waiting for feedback. maybe someone can help me.

wlan

03:00.0 Network controller: Intel Corporation Ultimate N WiFi Link 5300

the wireless lan system crashes if the iwlagn module is loaded and then unloaded several times (‘rmmod iwlagn’ and ‘modprobe iwlagn’ several times). a stack trace can be found at [4]. i’ve filed a bug report at [5].

acpi

my old 2.6.34 configuration reported acpi evnts for volume down/ volume up, fn+f3, fn+f4 (maybe), fn+f8, fn+f9 and fn+f10 as well as for pwr button. with the fc13 kernel configuration i’ve experienced that some if not all acpi events were not sent. still have to find out why this happens.

if i find out why that happens, i’ll post some news on that.

links

[1] http://ubuntuforums.org/archive/index.php/t-1037819.html

[2] http://lastlog.de/misc/wordpress/snd_hda_intel_backtrace

[3] http://www.alsa-project.org/main/index.php/Help_To_Debug_Intel_HDA

[4] http://lastlog.de/misc/wordpress/iwlagn_backtrace

[5] http://bugzilla.intellinuxwireless.org

Jason Jones: Two Delta 1010 Config

Sun, 26 Sep 2010 22:47:35 +0000

Okay... So I own and operate a recording studio, and I've been using an M-Audio delta 1010 as my primary audio interface. This provides me 8 channels of simultaneous audio streams coming into my workstation. Up until about a month ago, this has been very sufficient.

About a month ago, I ordered another Delta 1010, and planned on using this site's instructions on installing it. Tonight I installed the 2nd delta 1010, and planned on a pretty arduous journey getting it to actually run.

It took me 1 hour from installation to getting 2 streams of audio recorded (1 from primary 1010, one from slave 1010). These are highlights as to what I did to get it working, in hopes of helping others accomplish it even quicker.

After plugging in my 2nd audio interface, These are important points to getting it working.

I used SPDIF for sync.

Figure out which interface is connected to which device. There will be two 1010 devices now (using aplay -l) and you need to know which is which in order to get the SPDIF clock to route correctly. I accomplished this by using the -c option in envy24control (found in alsa-tools). I opened up two instances of envy24control (one with -c2 and another with -c3 (I have 2 other superfluous audio devices for 0 and 1). After doing that, I plugged a preamp into interface 1, and another in interface 2 - got some mics hooked up, and watched the meters as I spoke into each. This will tell you which interface is which.

After figuring that out, choose the interface you wish to use as your main interface. In the other interface's envy24control instance, click the "hardware settings" tab, and then under the "master clock" section, choose "SPDIF".

****VERY IMPORTANT**** - Connect a SPDIF cable from your main interface's "O" SPDIF connector (on the card, not on breakout box) to your slave's "I" SPDIF connector. Gotta make sure you do this in the right order, or you won't get very far.

Create an .asoundrc file and put in the configuration you find below. Save this in your primary user's home directory.

start jack, and then start ardour. You shoul see **16** input connections now. I use the following command to start jack.

jackd -R -d alsa -C multi_capture -P multi_playback -r44100 -H

You're done!

That's really it. I can't believe it was that easy, but thanks goes to whoever built the "jrigg" site containing the config below. I would have never even considered buying a 2nd delta 1010 without that resource.

Here's my hardware / kernel setup.

Linux / gentoo

RT - Kernel 2.6.33 (from pro-audio overlay)

AMD Phenom 2 940 (quad-core)

4 Gigs of RAM

ATI Radeon HD 4850 video card (dual outs)

dual acer H233H monitors (works great with two monitors)

ASUS motherboard (M3A79-T DELUXE)

Couldn't be happier with the the way this has been running. I've been using ardour in a professional studio evironment now for a year. No data loss, very few crashes, and customers absolutely thrilled with the product. To hear samples of audio recorded with 100% Linux-based software, click here. All plugins are Linux-based as well. I don't have to use wine or windows for anything but auto-tuning.

Below is the configuration which can be found here as well (with the hw parameters tweaked to my sound-card set up)

# .asoundrc for two Delta 1010s
#
# Create virtual devices out of multiple soundcards.
# JACK will need MMAP_COMPLEX support to use this.
# ICE1712 chip has 12 capture channels and 10 playback channels.
# No. of channels in slaves must equal 12 for capture and 10 for playback
# otherwise "invalid argument" errors result.

pcm.multi_capture {
    type multi
    slaves.a.pcm hw:3
    slaves.a.channels 12
    slaves.b.pcm hw:2
    slaves.b.channels 12

# First 8 channels of first soundcard (capture)
     bindings.0.slave a
     bindings.0.channel 0
     bindings.1.slave a
     bindings.1.channel 1
     bindings.2.slave a
     bindings.2.channel 2
     bindings.3.slave a
     bindings.3.channel 3
     bindings.4.slave a
     bindings.4.channel 4
     bindings.5.slave a
     bindings.5.channel 5
     bindings.6.slave a
     bindings.6.channel 6
     bindings.7.slave a
     bindings.7.channel 7

# First 8 channels of second soundcard (capture)
     bindings.8.slave b
     bindings.8.channel 0
     bindings.9.slave b
     bindings.9.channel 1
     bindings.10.slave b
     bindings.10.channel 2
     bindings.11.slave b
     bindings.11.channel 3
     bindings.12.slave b
     bindings.12.channel 4
     bindings.13.slave b
     bindings.13.channel 5
     bindings.14.slave b
     bindings.14.channel 6
     bindings.15.slave b
     bindings.15.channel 7

# S/PDIF section. Uncomment bindings if required.

# S/PDIF first soundcard (capture)
     #bindings.16.slave a
     #bindings.16.channel 8
     #bindings.17.slave a
     #bindings.17.channel 9

# S/PDIF second soundcard (capture)
     #bindings.18.slave b
     #bindings.18.channel 8
     #bindings.19.slave b
     #bindings.19.channel 9
}

ctl.multi_capture {
    type hw
    card 3
}

pcm.multi_playback {
    type multi
    slaves.a.pcm hw:3
    slaves.a.channels 10
    slaves.b.pcm hw:2
    slaves.b.channels 10

# First 8 channels of first soundcard (playback)
     bindings.0.slave a
     bindings.0.channel 0
     bindings.1.slave a
     bindings.1.channel 1
     bindings.2.slave a
     bindings.2.channel 2
     bindings.3.slave a
     bindings.3.channel 3
     bindings.4.slave a
     bindings.4.channel 4
     bindings.5.slave a
     bindings.5.channel 5
     bindings.6.slave a
     bindings.6.channel 6
     bindings.7.slave a
     bindings.7.channel 7

# First 8 channels of second soundcard (playback)
     bindings.8.slave b
     bindings.8.channel 0
     bindings.9.slave b
     bindings.9.channel 1
     bindings.10.slave b
     bindings.10.channel 2
     bindings.11.slave b
     bindings.11.channel 3
     bindings.12.slave b
     bindings.12.channel 4
     bindings.13.slave b
     bindings.13.channel 5
     bindings.14.slave b
     bindings.14.channel 6
     bindings.15.slave b
     bindings.15.channel 7

# S/PDIF section. Uncomment bindings if required.

# S/PDIF first soundcard (playback)
     #bindings.16.slave a
     #bindings.16.channel 8
     #bindings.17.slave a
     #bindings.17.channel 9

# S/PDIF second soundcard (playback)
     #bindings.18.slave b
     #bindings.18.channel 8
     #bindings.19.slave b
     #bindings.19.channel 9
}

ctl.multi_playback {
    type hw
    card 3
}

Dion Moult: The GIMP metal wires and abstract background tutorial.

Sun, 26 Sep 2010 08:22:30 +0000

Every WIPUP release, an abstract art splash image is created to commemorate it. Whenever I create one of these my preferred tool of choice is The GIMP. Although in many ways The GIMP isn’t as "powerful" as Photoshop, I still manage to do stuff I’d like to do with it. My recent splash image used a few tricks that I will share here which may be useful to others. Before we begin, this is the thing we’re going to learn how to create:

The first step is the wires. Let’s start with a black background and white foreground. Create a curved line via the pen tool, select a circle brush of a nice thick size, Edit->Stroke Path, Stroke with a Paint Tool (Paintbrush, do not emulate brush dynamics), and we’d end up with something like this.

Duplicate this layer, because we’d need this shape twice – once for the segments of the wire, and another time for the wire itself that joins the segments together. The next step is to cut out the segments. Create a white rectangle which covers the height of a single segment in a new layer, duplicate the layer and move it down. Keep on doing this until you have filled the entire screen. To make it faster you can merge layers together then duplicate the merged layer. Here is an image to show what I mean:

When done, merge all of the horizontal stripes into a single layer, right click on the layer -> Alpha to selection, select one of your wire layers, invert the selection (ctrl-I) and press delete. You may now delete the horizontal stripes layer. Here’s what you should end up with:

The next step is to create the metal gradient on the wire. Duplicate your segments layer twice. So we will have three layers in total – segments, copy 1, and copy 2. Invert the original segments layer to make it black (it will be invisible against the black background, so I have made the background grey in my next picture). Gaussian blur copy1 and copy2. Gaussian blur (Filters->Blur->Gaussian Blur) one more than the other in order to set the light direction. The benefit of doing this rather than simply stroking with a gradient brush is that you can slightly shear the gaussian blurred layers to create a less uniform gradient, and thus more realistic gradient. Then move copy2 to the right, and copy1 to the left. Use the left and right arrow keys. We will end up with something like this (zoomed in and cropped):

Right click on your black original segments layer (underneath), and do Alpha to selection. Invert the selection, select copy1, and delete. Then select copy2, and delete as well, this wil give us our gradient as shown:

Now let’s go back to our very first white stroked layer (seen in the first screenshot). Right click on the layer -> Alpha to selection, then Select->Shrink, perhaps by 3 pixels, invert the selection, and delete. Here’s what we get:

Now repeat the steps above to create the gradient. Use less gaussian blur though as this wire is thinner. When done, we’ve made our wire. Duplicate it and reposition it as you like.

Use the same technique to create the smaller wires underneath, but when creating the gradients for those, perhaps only blur it by 1px. To quickly and easily make many variations of wires (for the smaller ones) you can use Filter->Distort->Ripple. Use a high period, and a low amplitude. Use the sine wavetype. Keep on duplicating them and you’ll end up with something like this:

The next step is to add lighting. There are several techniques to do this and unfortunately pictures don’t really show much so reading through carefully is recommended.

The first step is to decide where you want your lightsource to come from. Merge your smaller wires into a couple layers, and go into Filter->Light and Shadow->Lighting Effects. In the Light tab, create a point or directional light (if you want more dramatic lighting) and place it where your light source is. Don’t make the intensity too high or place it too close, otherwise you’ll end up with a completely white wire. Don’t forget, these smaller wires are underneath. Play around with the lighting effects section, but not too much. This should just be a minor lighting effect.

The next technique is to add a dropshadow to the upper two segmented wires. Merge each wire into a single layer, Alpha to Selection, create a new layer, and fill it with solid black. Gaussian blur it, move the layer below, and tada, you have a shadow. use the perspective tool slightly to give the shadow more realism as though the wire is moving towards and away from you. This is a very easy and precise way to make shadows for any purpose and sure beats the dropshadow plugin which IMO sucks. Notice how my two wires tangle, so make sure you delete the shadow where necessary (alpha-to-selection, delete).

Another extremely useful technique is to create a new layer, set the mode to overlay, and then use a large, fuzzy circle brush with black and brush over darker areas. Use a white brush for the highlights (remember to use this where you put your first lightsource). Start with a large brush, then slowly move towards smaller size brushes, especially when one wire tangles over another. This should be your main tool to create lighting and shadows.

For the thin wires, you might want to give it a sharp lighting. Just alpha to selection, create a new layer, and fill it in with black. Offset it to the left or right (depending on your lightsource) a few pixels (with the arrow key). Sometimes you don’t even need to gaussian blur this. This will create a very sharp shadow, similar to an emboss or embed effect.

Finally, create layer masks (right click on layer->add layer mask) on strategic layers (or on wires that cut abruptly), use a soft fuzzy circle brush to fade them out nicely. Careful not to use these on the top two main wires otherwise it’ll look very unrealistic. Use black fade-to-transparent linear gradients at the top and bottom of 70% opacity to allow them to fade out slightly.

After all of this, here’s a possible outcome – but you’d have to really use your artistic sense at this point:

Well, the next step is to create the background. This is actually quite easy. Just create a linear gradient from one colour to another (I chose sky blue and pastel green), from the bottom to the top. Then do Filters->Render->Nature->Flame. In the Rendering tab, increase the brightness slightly, as too bright will make it too sharp. Your results will vary, so keep on trying gradients until you get one you like. In the Camera tab, change the zoom and X and Y values until it focuses on an area which you like. Render it, and do a few more until the entire screen is covered. Here’s what you might end up with:

Duplicate the flame layers and set the layer mode to Screen. This’ll give you a nice soft glow. You might notice that it clashes too much with the wires. So create a new layer, set to mode Overlay, and use a big black fuzzy circle brush to brush underneath the wires. This’ll give a nice "shadow". Here’s what you end up with:

Now let’s add some sparkly stuff. It might be good to add them where you wanted your light source to be. Choose a soft white fuzzy circle brush, select "Apply Jitter" in the brush options, and brush over the area. You might need to undo and retry it several times until you get what you like. Change the brush size and brush towards the top and bottom to make it "fade out". To make it more interesting, add a gradient from the top to bottom of any colour you want, and set the layer mode to "Colour". I didn’t like how vivid this colour overlay turned out, so I added a layer mask and used a large jittery brush to make the colour fade out in patches. Here’s what we end up with:

… and that’s pretty much it! I added the WIPUP text as usual. I hope you liked it! Any suggestions would be welcome. You can see the final image featured in the WIPUP release notes here.

Dieter Plaetinck: An rss2email fork that sucks less

Sat, 25 Sep 2010 18:33:08 +0000

Rss2email is a great tool. I like getting all my news messages in my mailbox and using smtp to make the "news delivery" process more robust makes sense.
However, there are some things I didn't like about it so I made a github repo where I maintain an alternative version which (imho) contains several useful improvements, both for end users and for developers/downstreams.
Also, this was a nice opportunity for me to improve my python skills :)

Here is how it compares:

(most of the changes are only in the xdg branch, which is the one I use and test)

criterion	official version	my fork
code hosting/release process	"release" = tarball containing updated code corresponding to various fixes, and snapshots of other projects (dependencies). no version control, not even separate patches. no bug tracker.	git. upstream tracking branch, master branch (basic "good stuff" patches), XDG topic branch (my favorite). Does not include code from other projects, rather list dependencies. Github issue tracker
code style	dirty (extraneous whitespace, ^M characters, incorrect permissions, loads of bogus [whitespace] changes, ..). The python code itself seems nice though	clean
file storage & config	all in ~/.rss2email, list of feeds, email address and feeds state go into pickle file. commands like 'r2e add', 'r2e delete', 'r2e list' to manage feeds. feed ids change when feeds get deleted. 'r2e email' to manage email adress.	adherence to xdg basedir spec. list of feeds goes into plaintext file, so does email address. state in separate pickle file. removed all the [now pointless] commands.
Temporary disabling of feeds	no (if you remove a feed, you loose the state info)	yes. comment it out or remove it. state info won't be lost
installation/runtime	no Makefile. no reliance on $PATH, locking code in python. Distros are applying patches and/or using custom wrapper scripts	smarter wrapperscript that prevents multiple runs, so removed locking code from python. Has Makefile. Easy to package
logging/debugging	useful info messages hidden by default, "verbose mode" and error messages using print calls all over the place	useful messages on stdout. additional (info/warn/error/..) logging and debuglogging use python module (xdg compliant)
web ui	yes	no (I don't need it)
windows support	yes	no (unless somebody ports xdg to windows and updates the wrapper script)

Other then that, there are also some smaller fixes in various places.

I'm using my version "in production" and it works great for me so far.
I have contacted the author and told her about my changes, but no response yet.
For now, you can treat this as an alternative version that stands on it's own. I made an Arch rss2email-xdg-git package in the AUR.

Oh, and I'm liking python :) Pretty nice and powerful language.

Jason Jones: gentoo + jack + dbus = broken

Fri, 24 Sep 2010 21:59:43 +0000

Linux + gentoo + jack-audio-connection-kit + dbus use flag enabled = doesn't work any more.

Okay... Just a personal note to remember.

DO NOT COMPILE jack-audio-connection-kit WITH THE dbus FLAG ENABLED! IT REMOVES jackd WHICH IS WHAT qjackctl USES TO START THE JACK SERVER.

So... Unless I'm missing something, this is pretty much the status quo for me - because this has bitten me TWICE now. I want to upgrade JACK for my studio's DAW, and the dbus flag breaks qjackctl by removing the jackd binary.

So... Don't compile jack-audio-connection-kit with the dbus flag enabled.

(Why is that flag even there? Am I missing something?)

Joachim Schiele: QAbstractItemModel vs QStandardItemModel

Mon, 20 Sep 2010 20:35:58 +0000

lately i started working on a torrent implementation for the evopedia project [1]. that will include the qt torrent example [2] but this posting won’t be about any torrent related stuff.

first i thought about implementing a QAbstractItemModel but reading various examples i discovered the QStandardItemModel.

what i want to do

i have an item, represented via a Model and visualized by a a QTreeView. the model basically hosts the item(s) and several views could display that item(s) in different layouts but using exactly the same data.

when a user clicks an item a torrent client will start downloading it. the torrent client will update the state of the download by using the item’s textfield via the model.

this can be implemented either by using a QAbstractItemModel or a QStandardItemModel, so let’s see which one is easier to be used:

discussion: QAbstractItemModel

i’ve been using this model quite some times now and i really love it but it has one big limitation: every time you want to change an item (which is an item of the model) you have to use the model to do that. that means if an item want’s to change itself, you have to query the representative QModelIndex and then use this with the model:

QModelIndex* m = index(0,0,QModelIndex()); // query the first row and column and a top level item, by using a invalid QModelIndex

m->setData(…); // here we set the data

QModelIndex’es are used by views to access the data by using the model interface. this is all about MVC and abstraction.

CONCEPT: most of the time there is no need for items to be able to change themselves BUT in this case it is a very handy thing.

two ways of implementing this:

either the item holds the torrent client object as well but the torrent client object will update the text of the item using the model
or one could also have a Model and a list of torrent clients and then the torrent client accesses the item also via the QModelIndex but this would mean having another lists to take care of

i think either case is error prone, so let’s see how this could be done using a QStandardItemModel instead:

discussion: QStandardItemModel

in contrast using this model you don’t use QModelIndex’es at all. instead you use an abstraction: the QStandardItem. the fun part about a QStandardItem is that it is holding data which is accessed by all views BUT it can also be used to change itself.

this code has been taken from the Qt QStandardItemModel manual (but slightly extended):

QStandardItemModel model;

QStandardItem *parentItem = model.invisibleRootItem();

for (int i = 0; i < 4; ++i) {

QStandardItem *item = new QStandardItem(QString(“item %0″).arg(i));

parentItem->appendRow(item);

parentItem = item;

item->setText(“foobar”);

}

therefore it is very easy to extend a QStandardItem by inheriting from it. let’s call this new item a QExtendedStandardItem. within that class one can have a torrent client object which does all the work. if the state of the download changes from download started to download finished, the torrent client can simply use setText(“finished”);

another point worth mentioning is that the torrent object could also expand the QStandardItems with additional childs which could be used to list up/down rates as well as peers/seeders and other valuable parameters (can be seen in the above image).

summary

using a QStandardItem makes it very easy to implement the torrent client.

links

[1] http://evopedia.info/

[2] http://doc.qt.nokia.com/4.6/network-torrent.html

Joachim Schiele: gentoo user? only 2gb ram or less?

Sun, 19 Sep 2010 23:32:24 +0000

gentoo

when compiling all the used software before installing it one might find himself in this situation i’m face often:

the comiler uses a lot of ram and therefore all the current running applications (if not killed by the OOM killer already) are swapped to the swap partition or swap-file.

that means after the installation (when there is free ram, the swapped out stuff still is in the swap space, as it is not transfered automatically back). in this case i often type this command:

WARNING: if you do have full ram, don’t enter the command below as the OOM killer WILL kill programs you might NOT want to be terminated, continue only if you understand what you are doing!

list of all swap partitions:

# cat /proc/swaps

Filename Type Size Used Priority

/dev/sda3 partition 244139800 16 -1

now let’s put that stuff back into ram:

# swapoff /dev/sda3

still i might need the swap space later on, so let’s reenable it:

# swapon /dev/sda3

now all the open programs run fast again!

Joachim Schiele: no more crashes after resuming from pm-suspend II

Sun, 19 Sep 2010 23:20:09 +0000

the problem

since my last posting the ‘hp elitebook 8530w’ kept crashing quite often, when doing a resume after pm-suspend. i think i found a cure. disabling ‘compositing’ using:

kde – system settings – desktop – ‘enabled desktop effects’

made it work. but for some unknown reason there were these random crashes when resuming.

the fix

it is NOT enough to disable this in kde as for some reason, it sometimes got reenabled without my influence (i don’t have any clue why). so today i tried to find a way to make these crashes reproducable: with some good results.

the problem was definitely related to ‘compositing’ (or AIGLX?) so i remembered the times when the ‘compositing’ extension was disabled by default, maybe for good reason?

i disabled it by editing xorg.conf:

Section “Extensions”

Option ”Composite” “Disable”

EndSection

Section “ServerFlags”

Option ”AIGLX” “off”

EndSection

after a restart of xdm with this new settings i could not get it to crash as i did before. that is good news to me! maybe it’s time to test x11-drivers/nvidia-drivers-256.35 again!

right now i’m using =x11-drivers/nvidia-drivers-195.36.24 with 2.6.34-gentoo-r1 with this graphics card:

01:00.0 VGA compatible controller: nVidia Corporation G96M [Quadro FX 770M] (rev a1)

i also have 2d and 3d acceleration running. *yepee*

update

21.09.2010 this is so strange, i still receive regular crashes on ‘resume’ but maybe they are not directly related to the nvidia module. the problem is that this is very hard to debug as there is no graphical output nor any log files to start with …

Kevin Bowling: Software Freedom Day 2010 – Gentoo Linux

Sat, 18 Sep 2010 15:37:58 +0000

I’m doing a presentation/Q&A for Software Freedom Day 2010 in Charleston, SC. CSCLUG is putting on the event.

Software Freedom Day 2010 – Gentoo Linux Presentation – .odp / Open Office Impress
Software Freedom Day 2010 – Gentoo Linux PDF – .pdf / Portable Document Format

Here is the presentation. Feel free to use or modify this under the terms of CC-SA3.0.

Share this article:

2010 SpamAssasin Public Service Announcement If you run a public mail server, there is a...
One Small Step for QT, One Giant Leap for Free Software QT Software, under the graces of Nokia, has released the...
Getting Beautiful Fonts in Gentoo Linux Here’s my easy and modern guide to getting pleasant looking...

Brian Carper: iPad? More like iAd. Vertisements.

Fri, 17 Sep 2010 01:43:35 +0000

Via Slashdot, it seems soon you may be able to subscribe to newspapers on the iPad in the near future.

Sure. Why pay $10 for a paper copy of something when you can pay the same $10 for a likely-DRM'ed copy that can only be read on a $500 portable computer?

In all honesty though, instant delivery, lack of clutter, "take it anywhere", being able to archive issues indefinitely, text search... those features might be worth the money, if it was a really good newspaper/magazine.

But wait, there's more.

The Cupertino company has agreed to provide an opt-in function for subscribers to allow Apple to share with publishers their information, which includes vital data that news organizations use to attract advertisers, industry sources say.

...

While the leap into the digital tablet market comes with short-term problems for newspapers, the iPad and future tablets will provide a new digital palette for publications to create sophisticated and lucrative ads, said Needham & Co. analyst Charles Wolf.

"I would say it's a risk, but I would argue it's a short-term risk," Wolf said. "If you can put animation and multimedia into ads, that will greatly enhance reader views. I am certain of that."

So as I understand it, first I buy a $500 gadget. Then I pay for a newspaper subscription. Then a bunch of companies want me to give them personal information about myself, so they can share it amongst themselves. And then I have to view ads.

Animated ads.

The only thing better would be if the iPad also woke you up at 4AM and tried to sell you life insurance. Maybe Apple would let me install a free ad-blocker script for my news reader though. It is my hardware, after all... pfffft, yeah, I could't keep going with a straight face.

And thus my desire to get an iPad, kind-of sort-of building over the past couple of months, once again flatlines.

Dan Ballard: Lisp (SBCL + emacs + slime) on Hardened-ish Gentoo on Xen (take 2)

Thu, 16 Sep 2010 17:27:58 +0000

A while ago I tried with mixed success to get Lisp onto my Gentoo Hardened server. I had to go a binary only route and kind of stopped there not taking it any farther. Now, 2 years later, I need the full meal deal, lisp + emacs + slime, on my server, which is now a Xen VPS with as much hardening as I could get (much less kernel based hardening since it's the VPS's kernel). It was still too much for SBCL to compile in portage so here's what I did to get it all working.

So you need an out of tree binary copy of SBCL. Live with it. It works. The problem with going with out of tree software, especially for a language, is that what ever binary you get isn't supported and hasn't been tested against all the software in-tree. For instance I initially tried the newest version of SBCL (1.0.42) but ran into problems with portage's stable slime.
Ultimately I went with the closest I could get to portage's stable version. Portage has 1.0.19 marked as the most recent stable version so I went out and downloaded the binary of that version

$ wget  http://sourceforge.net/projects/sbcl/files/sbcl/1.0.19/sbcl-1.0.19-x86-linux-binary.tar.bz2/download
$ tar -xjf sbcl-1.0.19-x86-linux-binary.tar.bz2

So change into the directory and check out INSTALL. Basically installation is easy. Binary SBCL is configured around installing into /usr/local but that can be gotten around. So we'll go with a more traditional install into /usr

Note: My test box is a VPS with a Xen kernel not a hardened kernel so I didn't have any PaX problems, but my notes for the last time I tired this on a full hardened install mention that you need do disable some PaX features before SBCL will work:

$ paxctl -p -e -m -r -x -s " on src/runtime/sbcl

Install to /usr

# INSTALL_ROOT=/usr sh install.sh

Now SBCL is installed but it won't work because the binary is preconfigured to look for the core in /usr/local. So we'll borrow the gentoo SBCL config files to get that setup properly.

/etc/env.d/50sbcl

 
SBCL_HOME=/usr/lib/sbcl
SBCL_SOURCE_ROOT=/usr/lib/sbcl/src

# env-update

The above file and command set up the system environment variables to tell SBCL where it's really installed. Now is as good a time as and to 'source /etc/profile' to get those changes.

Now SBCL is installed and working, we need to let portage know that. There used to be a 'emerge --inject' method, but that's been deprecated in place of a new provides file

/etc/portage/profile/package.provided

 
dev-lisp/sbcl-1.0.19

Now portage knows about our SBCL so we can start installing things that depend on it like the rest of our tool chain

# emerge cl-asdf emacs slime -va

So now we have all the pieces, all they need is some gluing together. Again we'll borrow from the Gentoo SBCL files.

/etc/sbclrc

 
;;; The following is required if you want source location functions to
;;; work in SLIME, for example.
 
(setf (logical-pathname-translations "SYS")
    '(("SYS:SRC;**;*.*.*" #p"/usr/$(get_libdir)/sbcl/src/**/*.*")
          ("SYS:CONTRIB;**;*.*.*" #p"/usr/$(get_libdir)/sbcl/**/*.*")))
 
;;; Setup ASDF
(load "/etc/gentoo-init.lisp")

/etc/gentoo-init.lisp

 
(in-package #:cl-user)
#+(or sbcl ecl) (require :asdf)
#-(or sbcl ecl) (load #p"/usr/share/common-lisp/source/asdf/asdf.lisp")
(push #p"/usr/share/common-lisp/systems/" asdf:*central-registry*)
(asdf:oos 'asdf:load-op :asdf-binary-locations)
(setf asdf:*centralize-lisp-binaries* t)
(setf asdf:*source-to-target-mappings* '((#p"/usr/lib/sbcl/" nil) (#p"/usr/lib64/sbcl/" nil)))

Now everything should work. You just need to set up your emacs and slime

~/.emacs

 
; your SLIME directory
(add-to-list 'load-path "/usr/share/emacs/site-lisp/slime/")
; your Lisp system
(setq inferior-lisp-program "/usr/bin/sbcl")
(require 'slime)
(slime-setup)
 
(global-set-key (kbd "C-c C-q") 'slime-close-all-parens-in-sexp)

Now It's all glued together, give it a go

$ emacs
M-x slime

If you don't get any compilation errors you should be in emacs + slime.

And there you have it, SBCL emacs and slime on Gentoo Hardened.

Cavets

1) For some reason this approach adds some annoying extra text to vanilla SBCL start up that I can't seem to get rid of

$ sbcl
This is SBCL 1.0.19, an implementation of ANSI Common Lisp.
More information about SBCL is available at <http: //www.sbcl.org/>.

SBCL is free software, provided as is, with absolutely no warranty.
It is mostly in the public domain; some portions are provided under
BSD-style licenses.  See the CREDITS and COPYING files in the
distribution for more information.
; loading system definition from
; /usr/share/common-lisp/systems/asdf-binary-locations.asd into
; #<package "ASDF0">
; registering #<system ASDF-BINARY-LOCATIONS {AAF8F51}> as ASDF-BINARY-LOCATIONS
*

2) The system I tested this on is a VPS so the kernel is a Xen kernel, not a hardened kernel, so there may be additional complications on a full hardened install. Please let me know if you have any, and especially any working solutions.

Jürgen Geuter: Diaspora fail

Thu, 16 Sep 2010 08:16:02 +0000

So Diaspora, the distributed and free alternative to Facebook, has released their first “developer release“. And who would have guessed: It’s a fail.

Not cause it doesn’t work right, it’s the first developer release, that was to be expected. It fails cause they rely on a platform that most people will not be able to run on their webspace.

Their model was identi.ca, the distributed Twitter alternative, which does things right: It relies on a platform (PHP) that people can run on cheap webspace. It really allowed people without special environments to run it for themselves. Maybe they had to deactivate some features (like XMPP posting) but the general thing runs just like most webapps do: Copy the crap onto your webspace, run the installer and play with it.

Now you need to get rails running and use the fancy shmancy GEMS thingy to install dependencies. Flameeyes has already covered why GEMS (and other similar distribution mechanisms), the bottomline is: Yeah those might make sense if you are one one of those legacy platforms that have no proper package management but for people with a sane operation system it just complicates things.

What we will end up with is a bunch of servers with outdated broken ruby libraries that don’t get updated. Yeah, awesome.

I understand that people wanted to pick one of those quick development frameworks, I myself like them quite a lot, but they are exactly the wrong thing for this kind of app: WordPress is so successful cause it runs everywhere. And you gotta realize that those frameworks help you get a prototype running fast, but in the end you end up dropping most parts of the framework anyways cause of specific needs for features they don’t cover.

I understand that the diaspora guys were under a lot of pressure: They got a lot of money to develop a free and distributed facebook clone and now they have to show results. Rails might have helped them get scaffolding done quickly. But it limits their distribution massively. It was a bad choice (Django/Python would have been just as bad just to make that clear considering that I hate Ruby with a passion and love Python).

Feel free to Flattr this post at flattr.com, if you like it.

Sven Vermeulen: SELinux quicky

Tue, 14 Sep 2010 21:44:27 +0000

I’ve been using SELinux for a few days now (in permissive mode, just to get to know things) and have learned a few interesting commands (or other nice-to-know’s) for using SELinux. Since I’m going to forget those the moment all is running well, I’ll “document” them here ;-) I’m not going to talk about the -Z switches in ps or ls, that has been documented sufficiently on the Internet already.

With sesearch you can query through the loaded policy. For instance, you want to know why you can execute sudo as a user (and not just due to the DAC permissions):

~$ sesearch -s user_t -t sudo_exec_t -p execute -c file -A

Of course, this is only one of the three requirements for a transition from user_t to user_sudo_t, for that you still need process transition and entrypoint:

~$ sesearch -s user_t -t user_sudo_t -p transition -A
~$ sesearch -s user_sudo_t -t sudo_exec_t -p entrypoint -A

Now, sometimes you find a rule that you didn’t expect:

~$ sesearch -s user_t -t dmesg_exec_t -p execute -A
Found 1 semantic av rules:
  allow user_t application_exec_type : file { ioctl read getattr lock execute execute_no_trans open } ;

This is because dmesg_exec_t has the application_exec_type attribute set. You can see the list of types that have an attribute set (or vice versa) with seinfo:

(Show list of types that have the application_exec_type attribute)
~$ seinfo -aapplication_exec_type -x
(Show list of attributes given to the dmesg_exec_t type)
~$ seinfo -tdmesg_exec_t -x

Now, during my browsing through the SELinux activities on my system, I noticed that I could run /usr/sbin/dnsmasq as root, without generating an error in the avc log. Yet sesearch didn’t give any results. I’ve almost killed a few kittens by searching for possibilities (perhaps types with exec_type automatically have application_exec_type – not, or perhaps the domain transitions to another domain first without me knowing – not, I would see that the process runs as a different domain then, which wasn’t the case). Luckily, dgrift on #selinux gave me the hint of checking the dontaudit rules as well:

~$ sesearch --dontaudit -s sysadm_t -t dnsmasq_exec_t
...
   dontaudit sysadm_t exec_type : file { execute execute_no_trans } ;

So there we had it – it was being denied, just not logged. And because I ran in permissive mode, it gets executed anyhow. I disabled the dontaudit rules and got the avc denial I was expecting:

(Disable dontaudit rules)
~$ semodule -DB
(Reenable dontaudit rules)
~$ semodule -B

Matija Šuklje: HP Photosmart C4380 over WiFi problem with HPLIP and CUPS solved!

Mon, 13 Sep 2010 21:16:00 +0000

As I've already reported, I've had a very annoying problem of my HP Photosmart C4380 over WiFi not printing.

Now I finally got it to work!
Update: a better solution ...was my fault really.

The real problem seems to be that CUPS doesn't support pure Avahi yet. In fact support for Avahi (and Bonjour over it) is planned for the 1.5 branch, but only mDNSresponder. Meaning that CUPS and HPLIP just couldn't understand eachother where the printer is located. Althought Zeroconf (Avahi being its most prominent and FOSS solution) is a great thing, it's pretty annoying when you get used to it and it fails ...in my case: forgetting to run avahi-daemon.

So there's two solution now.

Avahi with mDNSresponder support

After Jason reminded me in the comments about Avahi working, I started playing around a bit and, yes it works.

What's more, I got it working without editing any config files.

What I forgot is that you need to have to run Avahi daemon in order for it to work.

So with a simple:

/etc/init.d/avahi-daemon start
rc-update add avahi-daemon default

it works.

Since HPLIP uses mDNSresponder you can either:

install Avahi with USE="mdnsresponder-compat"; or
for a pure Avahi system see if Jason's way works.

Using SLP instead of Zeroconf/Avahi

This trick is pretty simple — use static IP instead.

What I did was add the printer as normal with hp-setup and then in CUPS' web interface changed the Connection URI to hp:/net/Photosmart_C4380_series?ip=192.168.1.101. Just change the IP accordingly. You can get your printer's IP by different methods — with hp-probe, on your router, or by accessing the settings from the small LCD of the printer itself. I imagine people who use the hp-setup GUI can just select the SLP discovery method to do the same automatically.

~~I'm still not 100% happy with the fact that it happened, and am not sure that was the only problem I had ...but at least it works again and Avahi support is planned.~~

hook out >> off to bed...

Luca Gasperini: Gnome places double entries

Mon, 13 Sep 2010 18:42:17 +0000

As a gentoo user I’ve been stuck with gnome 2.26 for ages and when I emerged gnome 2.28 I found out that two out of three nfs shares had double entries in the gnome places menu. It seems that there is some sort of problem with gvfs, some users have reported double entries in gnome places for local mounts and for remote shares.
I could solve the problem (now, end life of gnome 2.30) by mounting my shares with “defaults” as an option. These things make me wonder why I like computers.

Nick Cunningham: Damn you mod_security!

Mon, 13 Sep 2010 13:44:18 +0000

So I was attempting to write this series of SSL guides last week when I hit upon a problem, after playing around with Cygwin for a while I decided to update the first part of the series to use a different path to store the certificates as it made more sense and would be pretty

Jason Jones: 3 Nephi : 14

Mon, 13 Sep 2010 09:08:29 +0000

The golden rule seems to have come out of this chapter.

The first thing that caught my attention in this chapter is the possible meaning of verses 4 and 5. the "mote in thy brother's eye" verses. Some people would interpret this as "live and let live", or "worry about yourself, not others". To me, this would be the case if the last sentence of verse 5 wasn't written.

"...and then shalt thou see clearly to cast the mote out of thy brother's eye."

The meaning of the beginning of this chapter seems to hinge on verse 1, which says "Judge not". If that were the only thing written, then I wouldn't have a problem with it. However, it goes on for 5 verses about the details.

Like it or not, we're almost always casting judgements. Using an extreme case, if someone is pointing a gun at my wife's head, my judgement is that their intent is to shoot her. Only a fool would judge otherwise. I'm not going to judge that a drunk person is going to be able to make rational choices, nor am I going to judge a mentally unstable person with a weapon as safe. We always make judgements.

We've just got to be careful about the associations made within us that form our judgements. If we judge harshly, then the consequence of our harsh judgement seems to be a harsh judgement of God against us.

We also should be concerned about purifying ourselves, instead of only worrying about the flaws of others. The flip-side to that is, we also shouldn't be uncaring for others. If we are actively doing our best to be our best selves, then it is given us a more clear view of what others are in need of. It wouldn't be doing us any good to ignore that. Gentile persuasion is the Lord's way, and while we are in pursuit of self-cleansing, we should use the Spirit to see how we can help others do the same.

That's all for now, but so far, we've only touched the first 5 verses. This whole chapter is all about how to more properly govern ourselves, as well as how much God loves us.

Sven Vermeulen: Switching to hardened

Sun, 12 Sep 2010 11:41:49 +0000

Yesterday (and this night) I successfully converted my system to a Gentoo Hardened system. In my case, this currently means that PaX has been enabled and I am currently running the system (which is an x86_64 laptop) with SELinux in permissive mode (so it won’t enforce the policies yet, but report violations so I can see in my logs if enforcement is possible or not). The permissive mode will be on for quite some time I would assume, as getting SELinux active on the system involved quite a few ~amd64 packages and I’m not too fond of using that branch (I’m more of a stability guy).

Dan Ballard: Google’s Peter Norvig coming to UBC! I’m excited!

Sat, 11 Sep 2010 18:55:37 +0000

From UBC CSSS News letter:

1. Distinguished Lecture Series*
Speaker: *Peter Norvig, Director of Research, Google*
Date: Thursday, September 23
Time: 3:30 - 4:50 pm
Location: DMP 110

For those you you who forget, Peter Norvig is Google's head of R&D and wrote "Paradigms of Artificial Intelligence Programming: Case Studies In Common Lisp" which I am slowly working through. I'm excited!

Luca Gasperini: Move files in bash with tar and progress bar

Sat, 11 Sep 2010 13:14:57 +0000

Example
gas@caldaia ~ $ my_move source_dir/ dest_dir/ 43.5GB at 61.0MB/s eta: 1:36:59 11% [===== ]

You need to install “command line progress bar”, in gentoo it can be installed with emerge bar in ubuntu the deb can be grabbed here and installed with gdebi.

my_move:
#! /bin/sh SOURCE=$1 DEST=$2 SIZE=$(du -sb $SOURCE|awk '{print $1}') (cd $SOURCE && tar -cf - *)|bar -s $SIZE| (cd $DEST && tar -xBpf -)

Steven Oliver: Git Pains

Fri, 10 Sep 2010 21:58:30 +0000

I don’t currently own a linux PC. My old one crashed. Never bought a new one. I currently do all my fun open source stuff on a Mac laptop (10.6.4 if you care). The only problem with that is almost all of the neat open source stuff you find on the web is for Linux. Including the help documentation and such. Which works a lot of times, Mac is UNIX based after all, but then there are always those cases where it doesn’t.

I originally installed Git on my laptop when it was Git version 1.5.x. They’re currently rocking 1.7.x. Well needless to say a lot has changed; including apparently where they install the stupid program at. I have been running into all kinds of issues with Git recently. Craziness like commands straight from the Git documentation online not working. Stuff like,

git remote add --mirror ....

Yeah. That should work. It’s documented. Well it didn’t. So I downloaded the latest version of Git and installed it. Then randomly I did a git -v. It still said 1.5.x!! Here’s how I fixed it.

sudo rm -rf /usr/bin/git*
sudo rm -rf /usr/local/bin/git*
sudo rm -rf /usr/local/git
sudo rm /etc/paths.d/git
sudo rm /etc/manpaths.d/git

Then I reinstalled git. Hope that helps someone!

Enjoy the Penguins!

Jürgen Geuter: Time to say something nice about an old “enemy”

Fri, 10 Sep 2010 21:12:51 +0000

I think it’s time to say something nice for once about a company that we’ve all spend quite some time on hating: Microsoft. We all know the old drill. Microsoft is evil, they spread FUD and are generally the enemy, but I feel that lately there has been some change.

Not only do they offer a substancially better patent grant that SUN did for JAVA but their Codeplex group recently donated 25,000$ to the Mercurial project, and they have code in the Linux Kernel for their Hyper-V Virtualizer.

They haven’t pulled around completely and I don’t think that all of their departments have really dropped the whole “hate FLOSS” attitude, but I think that we have reached a point where someone else has taken the cake as the greatest asshat in town: ORACLE.

They bought SUN and didn’t waste a lot of time: They sued Google over JAVA patents, killed the Opensolaris project and I guess they’ll drop OpenOffice soon like a hot potato.

Their little change does not make me want to use their operating system (I have a console for gaming I don’t game on PCs and for work I need something serious) but I think that they have done some good lately and I like seeing that. There is room for proprietary companies, it’s not a binary thing for most of us, we just want all people to play fair and get along, if Microsoft is really getting their shit together I’m happy to welcome them.

Feel free to Flattr this post at flattr.com, if you like it.