Rundown of changes:

• Detection of Linux Mint (LP:#971643).
• Minor documentation fixes (LP:#944371).
• An option to disable countdown window (LP:#934586).
• Start/Stop options for indicator menu (LP:#934841).
• More resizing handles on region selection window (LP:#946403).

Few other additional features:

• Added parameter –nosound to disable Pulse Audio and sound detection.
• Added parameter –test to generate a test video signal instead of capturing screen.
• Basic Pulse Audio check added if there are no PA libs, sound will be disabled.
• GtkBuilder UI (again), replaced GtkCheckBox with GtkSwitch.
• Removed Global Menu and replaced About option with a button (Contribution by Georgi).
• Blinking indicator icon.
• Added ‘Cancel coundown’ option to the indicator menu.
• Speed optimizations for VP8 and H264.
• Changed H264 container from Matroska to MP4.
• Added streaming options to H264.
• … perhaps there are more.

You can get Kazam at the usual location on Launchpad.

Python-keybinder still didn’t receive an upgrade, so I was forced to seek another way. After some googling I figured that Xlib provides everything I need. Awesome!

Not really. I made a working python example and when I transfered the code to Kazam everything blew up. It is not really smart to mix Xlib and Gtk. I started digging again. So, keybinder used Gdk for capturing all those keys.

I can do that! Right? Well, I can’t really. Gdk_window_add_filter() wasn’t included in Gnome introspection for some reason. Why? So far nobody knows.

Until I figure this out and get someone to do some work on introspection, Kazam will be left without global shortcuts. I hope this time will be measured in days, not weeks. :)

Call for testing and an open season for bug hunt. We will be hunting bugs in Kazam Screencaster and Gwibber Microblogging client. To help with the testing you will need the latest version of Ubuntu 12.04 – Precise Pangolin with Gwibber 3.4 (3.3.90) and Kazam 1.0.4 (1.0.4.1 NCC-1701E) installed. Gwibber testing will be exclusive to Precise Pangolin, Kazam testing will also cover Oneiric Ocelot. You will have to manually install Kazam by running the following command:

$ sudo apt-get install kazam

Instructions for Gwibber manual tests are available here. Instructions for Kazam will hopefully be published by tomorrow.

Everyone who would like to participate in this testing event is welcome to join #gwibber and #kazam on irc.freenode.net where event will be coordinated. For questions, contact BigWhale.

Bugs are reported in Launchpad, https://bugs.launchpad.net/kazam and https://bugs.launchpad.net/gwibber.

Many thanks to Kiberpipa, they provided us with a place  to stay and their infrastructure. Slovenian LoCo Team will be live streaming the whole event and we will have a Google Hangout setup so anyone can join.

Good hunting and keep in mind that we are also accepting new feature requests. :)

Kazam Screencaster reached certain maturity and I am happy to announce a stable 1.20 release. Codenamed as NCC-1701D it hit PPA repositories just yesterday.

For Ubuntu 11.10 Oneiric Ocelot and Ubuntu 12.04 Precise Pangolin you can use a stable PPA repository:

$ sudo add-apt-repository ppa:kazam-team/stable-series
$ sudo apt-get update
$ sudo apt-get install kazam

Ubuntu packages were also tested on Linux Mint and you shouldn’t have any problems running Kazam on Linux Mint 12. I also tested Kazam on Fedora 16, everything should work as expected, but you will have to solve all the dependencies by yourself, because I cannot provide a working RPM package.

All of the issues reported so far were fixed. A short list of smaller new things that were added:

• region size is now displayed in the region window,
• screen resolution change is detected and video source selector is now updated accordingly,
• debug option –debug was added it should be used when reporting bugs, expect a lot of debug output,
• countdown window and about dialog box made prettier,
• App indicator menu and Gtk Status Icon were slightly changed,
• various changes in packaging, some RPM info added to setup.py

For the end, the latest screen shot from Precise Pangolin and a rumor that Kazam was uploaded to Ubuntu Universe repository and is waiting for review.

Kazam Screencaster 1.2.0

Please report all the bugs, wishes, suggestions, and feature requests on Launchpad.

Custom region recording.

Kazam 0.91-ubuntu5 and ubuntu6 for Oneiric Ocelot and Precise Pangolin packages are available. Source code is acab be found in the standard place on Launchpad. This is the last planned release before 0.92 which will be a week before Precise Pangoling Feature Freeze. Right now there are no immediate plans for adding new features before 1.0 release. Kazam release will coincide with Precise Pangolin release. So, a quick rundown of new feature in this release:

• Gtk3 port is finally complete.
• Appindicator will now fall back to Gtk.StatusIcon.
• It is now possible to record custom screen region.
• Recording frame rate is now a customizable option from 1 to 60 frames per second.
• If not needed or wanted, mouse cursor can be turned of in the recording.

You can get the packages from the unstable ppa. And please, do report those bugs you find. :)

Bugs addressed and fixes commited to: #896806, #661767, #656864, #662374, #635921, #732166, #739171, #908768, #660673, #660687.

If you had Kazam previously installed from unstable series PPA it will be upgraded automatically. If you are trying new installation, then you’ll find more detailed instructions here.

UPDATE: Kazam 0.91 was released, read more about it.

UPDATE: Kazam 0.91 was released, read more about it.

Main Window

Luckily there is not much more than that. ;) Here’s how it looks now and this is probably how it is going to stay.

Main window with the new options.

You will notice two selectors for audio source. Recording of multiple channels is the very first thing that will get implemented after the transition to Gtk3 is complete. Now if you enable recording from first audio source then second audio source is also enabled and it will also be recorded.

Encoder type will, for now, offer only GStreamer VP8/WebM and GStreamer H264/Matroska. In the future I am planning to add support for HuffYUV format, to really get that lossless quality. Also, ffmpeg is still not completely off the table and it might return in the future.

Few minor changes are also in the behavior of main window and application indicator. Indicator is now always present and an icon will give clear indication of what is going on with Kazam. Status can be: stopped, recording or paused. When recording and paused main window will disappear. Upon ending the recording and finishing with the export, main window will open again.

Other stuff

Countdown splash screen required almost a complete rewrite. Gtk3 is handling images and transparency a little different than Gtk2 and there was a reported bug (lp:#662374) about countdown window being fixed size which interfered with translated strings. Together with that I also added an option to increase or decrease countdown, closing another wishlist bug (lp:#732166).

Screens are now a little easier to identify if you have different resolutions, I am thinking about “Identify Screens” button that could display two identification windows for a short period of time on each screen.

I’ll try to squeeze in lp:#661763 and I’ll take a good look at lp:#910654.

Next release

Codename for this first release is NCC-1701A with the version number 0.91. Scheduled for February 1st, just a day before Precise Pangolin Alpha 2. It should be feature complete and UI complete, but I’ll leave some room until feature freeze in case something pops up. Most of the planned fixes are already committed in the unstable branch. Gtk3 port is being developed in my personal separate branch.

Screen selection and multiple audio channels.

Availability of audio channels.

Feedback

Is appreciated. Here, in launchpad, on IRC (irc.freenode.net #kazam) or in the mail.

Later, I’ve exchanged couple of emails with Andrew and offered my help. So, I’ll release an official version of Kazam in mid January and continue to work on it.

UPDATE: Kazam 0.91 was released, read more about it.

What has changed?

There were many isses with ffmpeg from audio recording device selection to codec nightmares. What to install, where to get it? Compiling my own codecs?! Seriously? Truth be told, ffmpeg and H264 are a minor nightmare on Ubuntu there are problems with legal restrictions and there are problems with continuously changing parameters for x264 and ffmpeg, beside that, ffmpeg has API that could be used for accessing it.

The first thing to go was ffmpeg support, then I decided to drop H264 encoding too. I sacrificed some quality and CPU power for ease of use. Ffmpeg is still there, just disabled and waiting for better times in one of next releases I’ll enable it for those users that have necessary support libraries installed and I’ll provide a short guide on how to install them.

• Default backend is now gstreamer.
• Video is encoded with vp8 instead of x264.
• Audio is encoded with Ogg Vorbis.
• Audio/Video container is now WebM.
• Package dependencies revised to include only essential packages.

Pulseaudio support

There as no real support for selecting audio devices and which input to use for recording sound. Right now the default pulse audio input was used and user had to change the default in order to use different input. To make things worse, Ubuntu sound settings don’t list monitor devices as input devices and recording application sound was close to impossible for most of the users. After a lot of digging I managed to whip up python support for pulseaudio and extracted list of all audio devices that can be used for recording.

• Pulseaudio basic support.
• Independent audio input device selection, no need to open sound settings.
• Volume setting slider in the working.
• The path to multiple source recording is now open. Next release should support recoding audio on two channels (application sounds and voice-over commentary, for example).

Kazam Screencaster - Audio selection

User interface changes

In Ubuntu 11.10 I found menu bar icons too subtle and hard to read. I changed them just tiny little bit and I added colors to record and pause indicator. Before I got rid of ffmpeg I also added one combo box for selecting back-end, currently it doesn’t do a thing.

• Panel icons made a bit less subtle so that you can easily notice if you are recording or not.
• Back-end selection combo box.
• Audio source selection combo box that actually works.
• Temporary files are now removed if you quit application and application exits gracefully.
• Added prefix “kazam_” to temporary files for easier recognition.

Kazam Screencaster - Combobox too long?

Is something broken?

Most likely it is. I didn’t even try to edit any of recorded videos. I will do that at some other time. Those parts of code are still as they used to be and rely heavily on ffmpeg. If you don’t have ffmpeg installed Kazam will break if you try to edit recorded captures.

There’s also multiple screen support. In my configuration it works, two screens side by side. If you have more than one screen not side by side or you have multiple screens then your mileage my vary. Gstreamer is very precise when it comes to specifying what part of the desktop to record.

How to install?

Update (01/01/2012): Kazam Team PPA from the unstable series is already available, info below.

Source code and PPA build for Ubuntu 11.10 Oneiric Ocelot is available in my personal PPA. Official PPA will be available after some more polishing and few minor changes, most likely in January 2012.

Source code is available in Launchpad and there is an official unstable PPA available. Version number was bumped to 0.13.

$ sudo add-apt-repository ppa:kazam-team/unstable-series
$ sudo apt-get update
$ sudo apt-get install kazam

All the input is greatly appreciated!

“Fantasy Flight Games is delighted to announce the upcoming release of Merchant of Venus, a board game of space traders and alien riches. Two to four players race through the galaxy on a wild dash for cash, locating and establishing vital trade lines to transport goods from the planets where they’re common, to the planets where they can sell for maximum profit.”

Then on the 22nd of October Stronghold games posted the following announcement on their web site:

“Stronghold Games, in conjunction with the Designer and license holder, A. Richard Hamblen, announced yesterday that we will reprint “Richard Hamblen’s Merchant of Venus“. We look forward to bringing you this great game during 2012 !”

If you care about this publishing fiasco then read on …

Naturally we don’t own Merchant of Venus. For all of you who were born after or around the year 1980; it is a classic science fiction board game with beautiful artwork and unprecedented rules of play. Trading in space while equipping your space faring bucket with lasers, shields, engines, additional cargo space … we are thinking about Elite, how about you? Perhaps Privateer if you aren’t as ancient as we are. Anyway, we will leave the review for a another post. The game was published by Avalon Hill; designed by Richard Hamblen and it all happened in year 1988.

The game has been out of print for some time and we would love to get our hands on a copy but eBay prices are to steep for our pocket. Both announcements gave us some positive love towards owning a copy; but on the other hand they generated a lot of fear that we will have to wait for some time due to possible lawsuits that both FFG and Stronghold Games could face.

In the light of developing events Stronghold Games clarified that they have come to an agreement directly with Mr. Hamblen for a license¹ to reprint the original version of Merchant of Venus board game. It seems that Stronghold Games and Mr. Hamblen believe they have exclusive license over the game. Unfortunately FFG also bought the exclusive license from Hasbro and they also believe they have the exclusive thing. Hasbro owns Avalon Hill company and all their licenses, which probably include the original Merchant of Venus license since they already released the game in 1988. Fantasy Flight Games CEO, Christian T. Petersen, posted on the BoardGameGeek forums his side of the story. We are wondering who got the most $$$ from the whole shady deal? Right now we known that the only one who will surely gain is Mr. Hamblen. It seems that he knew about both deals and he is turning out to be the master manipulator in the whole situation.

Both companies claim to have license to publish this long lost marvel. We would prefer that only one company prevails. We really don’t care if it is Stronghold Games or Fantasy Flight Games as long as Merchant of Venus gets published in due course. We believe that two versions of the same game would create a chasm in the market and divide player base. For now all parties seem to have valid claim to publish the game and the question which arises is, which side to back up?

Backing your side

There are some arguments which tell that Stronghold Games will make more accurate game. For example, they will do a reprint of the original game with support for six players while Fantasy Flight Games is thinking ahead and they are planning to release a four players base game, hinting to further expansions.

Then, there are actual session reports which tell us that the game of four players is an optimal group size for the game in terms of game time, components used and optimal fun. Fantasy Flight Games will surely do an expansion at a later stage of development allowing 5-6 player game. It would be the natural course of action.

Information available about rule set is scarce, but enough to push the original fans toward Stronghold Games. Fantasy Flight Games already announced changes to the original rules. It is unclear at this point if the additional rules will be optional or required.

From the manufacturing standpoint we feel much better with Fantasy Flight Games since they have an enormous amount of resources in designers, developers and money which they can throw in the production of the game. Stronghold Games just does not give us the same vibe.

While religious wars are being fought on the BoardGameGeek forums we feel that it is time for the gamers to decide for themselves what is best for them. Forum wars will clean all the dirty laundry of both companies and all fans who feel the urge to raise voice can speak up. The real losers here will, again, be the players. For now, let’s wait for the lawyers to get to the bottom of this licensing issue, stop the speculation and start to plan the strategy for your victory.

I can not wait for the Venus to rise

If you find the game interesting and can not wait for the war to end then go to BoardGameGeek site search for board game Merchant of Venus² and in files section search for files from user Dathkadan. He has recreated the game art with expansions. Print the game components, and recreate your own copy of this old gem.

What users in general need and want:

• stability
• reliability
• security
• simplicity

In reality you always need to keep those properties in some sort of balance.

What is a rolling release distribution?

In simple terms this means that your software is constantly updated and upgraded. You usually get all the new features and bug fixes very quickly and your operating system doesn’t have a distinct release number. If you installed FooBar Distro 1.5 today it will be 1.5 only today. In a couple of weeks you will get new browser, new office suite, new email software and so on and so on. Then at some point in time developers of FooBar Distro decide that current snapshot of all the software will be called FooBarDistro 2.0 and you don’t need to upgrade anything, you’re already there. You migrated slowly from 1.5 to 2.0.

The upside is that you are always running up to date software with all the bells and whistles. The downside is well, where do I begin?

• The need to upgrade. You must upgrade if you want security fixes. When I was using Gentoo there was no option to get only security fixes, you had to install new software. If I didn’t update for a longer period of time an upgrade usually broke during the upgrade process.
• Maintainability. If you have a lot of machines to maintain a rolling upgrade that can break becomes a pain. Even if you know how to fix it, it will require more time than a simple security update and bugfix release.
• In-house support. You will live in a dependency hell if your third-party software requires different versions of some essential library. Which immediately brings us to the most important downside of rolling releases.
• Third-party support. Developers and OEMs are having difficulty keeping track with current Ubuntu release cycle. New version every six months is a pain to support and support matrix for a product will look like a crosswords puzzle after a while. If you have libraries literally changing on a daily basis you are doomed and any kind of support becomes questionable.

For some people six months isn’t often enough and they would like to see a rolling release Ubuntu. Those people are a minority and I think that the correct approach is to have an even longer release cycle.

Why a longer cycle?

Because of third-party developers. Hardware or software development is a lengthy process in which you design and develop a product that will cover as much users as possible. You want a large user base. If you need to support a product for various operating systems with various libraries, software and architecture this is all what you will do. Support it and you will have no time to develop new features. Commercial software developers should already be targeting only LTS releases with their products, which is quite hard when majority of users will switch to a new Ubuntu release when it becomes available.

Ubuntu should move to a 12 month release cycle, with this they will give more time to their own developers and to all the contributors and third-party developers. What about new applications? Office suite? Browser? We would keep those in official PPA repositories that could be updated more frequently and independently. Applications that aren’t tied to the system can this way have more regular upgrades.

There are a couple of downsides to this solution. One is that someone will have to maintain those PPA repositories. Which shouldn’t be such a problem, but any additional work takes some time from the actual development. Second problem could arise with the upstream components. A year is a long time, what if upstream releases new version couple of months after the Ubuntu release? With six month release cycle this will amount to only four months of wait for early adopters. With a longer cycle this means ten months of waiting. For some this could be unacceptable.

Benefits of a longer cycle:

• More development time. This is critical, every release of Ubuntu that introduced something new on a bigger scale (PulseAudio, Unity, Gnome3, …) always seemed like it is a rushed-out release and the next release was a bug fix release.
• More testing time. Alpha and Beta cycles are not long enough. Ubuntu is still pretty much dependent in the community out there. Community needs more time to do all the testing. I do not know how exactly Canonical manages their QA but before I decided to be more active with Ubuntu development, I rarely tested Alpha and Beta releases. I usually didn’t have enough time to do it.
• Graphics cards vendors will have more stable environment to develop drivers for. Every Xorg release breaks proprietary graphics drivers and it happens every six months. Now it would happen only once a year.
•  Less demands for backports. Because of less releases there would be less backports, backports that could break and wreak havoc.
• Lastly, again, more time. More time for users to adapt to new release, more time to test it and more time to enjoy it before going through the frustration of yet another upgrade.

Longer release cycle will bring more stability and reliability. Security can also improve because new and untested software is introduced less frequently and for the same reason usability can also be increased. Learning all the quirks of the new software is a pain and users don’t like to adapt too much. Granted, in the long run everyone adapts, but from the UX perspective, changes shouldn’t be made just for the sake of changes and not too often. Even one year period could be seen as too short for big UI changes.

Good user interface concepts lasted for years before they evolved and were changed. Competition is a fine example of that.