hackergotchi

Talking about Folks at the Desktop Summit

Saturday at 11:20 in Rm2002, I'll be giving a talk on Folks, covering exciting topics like:

  • What is Folks?

  • What isn't Folks?

  • Folks in Gnome 3.2, MeeGo, and Qt Applications

  • How Folks can help you be a more-complete person*

A lot has been going on in Folks lately (including the 0.6.0 release, which will be happening shortly), so you really won't want to miss it!

I'm going to the Desktop Summit

*This may or may not require subterfuge
hackergotchi

IM, Contacts, Social hackfest successful

I just got back from the highly-productive and thoroughly-enjoyable IM, Contacts & Social hackfest.

We discussed a number of issues related to chat, video/audio calls, contacts, and social web service integration into Gnome. After each session, we filed all the necessary bugs to make sure we don't lose track of anything. Check out the session pages for notes and links to the related (mostly new) bugs.

It was especially good to have usability people Nick Richards and Allan Day on hand for our user-facing discussions. In general, I'm very happy to see Gnome continuing to move toward an interaction-design-driven development methodology. I'm quite confident that having a strong, consistent design-first, code-second process is the way we can (more than) close the usability gap that all-too-often exists between open source projects and their closed source counterparts.

Gnome 3.2 is going to have fantastic chat and contacts integration with the appearance of Gnome Contacts, Folks getting an EDS backend, Folks/Gnome Shell integration, and a number of other integration points. And it will only get better from there!

I would like to thank my employer, Collabora, for sponsoring the hackfest with a venue and especially the time of so many relevant developers (and to Guillaume for organizing the hackfest). Thanks also to the Gnome Foundation for additional sponsorship and Intel and Red Hat for sending along a few people each. This type of collaboration will help ensure that Gnome (and related platforms) not only keep up with our closed-source friends but begin to surpass them in ways we traditionally haven't (particularly usability).
hackergotchi

Updated Folks/QtFolks roadmap

We've focused the plans for Folks for the next few months and beyond in our much-revised Roadmap.

Some highlights:

  • Finishing the Evolution Data Server backend and making it the default primary backend

  • Adding search-based contact retrieval

  • Lazy-loading of attributes (to avoid a performance crunch as we've added a lot of attributes)

The Tracker backend will be the default primary backend for MeeGo until it officially switches to EDS for local contact storage. Around that time, we'll spin it off into its own module to make sure Folks can play well with out-of-tree backends.
hackergotchi

New features in Folks 0.4/0.5

A lot has been going on in the world of Folks lately, so it's time for another update. There have been a ton of fixes and clean-ups in the code (including some well-documented API breaks). We're narrowing down on another set of API breaks in our unstable 0.5.x branch before starting the stable 0.6.x branch. Our goal is to have very few (if any) API breaks after that.

Perhaps more importantly, we've added a lot of great features in the last few months. Read on for more (heavily-summarized) details.

I'd like to thank Intel for both sponsoring most of this work and understanding open source development. We were able to work in the open very early on and keeping things upstream was a goal, not a challenge. All the work is LGPL 2 or 2.1 and none of it requires copyright assignment.

New Fields

One big focus in 0.4.0 was adding a number of address-book-type fields, like:

• Role (within an Organization)
• Birthday
• Notes
• Postal address

In all, contacts from Folks support most of the generally-useful fields you'd find in a typical address book.

The remaining features are new as of version 0.5.0.

Primary backend

A design goal of Folks, since the beginning, has been to simplify the way writes happen. There is, at most, one (primary) backend which provides contact linking hints to the aggregator and which stores address-book-like contacts (which are generally fully-writable for all supported fields). This has historically been (hard-coded as) our simple key-file backend but it is now configurable.

Tracker backend

The new Tracker backend acts as an address book and supports both reads and writes. Creating this help fill out our support for local, address-book-like contacts, so that portion of Folks is in decent shape now. It can be set as the primary backend, though the default remains the key-file backend. My colleague, Raúl, did a great job putting this and a sizable set of tests together.

libsocialweb backend

We've also added a backend for libsocialweb which lets us expose contacts from various web services through Folks. This required a little work on libsocialweb itself to add a Contacts interface and implement it for a handful of services (so far, Facebook, Flickr, Twitter, and Last.FM). Alban picked up on Marco's work in this area, added the last three services and implemented the newer Folks interfaces for the libsocialweb backend.

I think this area is especially exciting, since it lets us include contacts' details from a very wide range of services in the future, just by adding more services to libsocialweb.

Potential Matches

A long-requested feature has been the ability to hint matches between various contacts within Folks (since it's tedious to have to manually discover them all yourself). This clears the way for us to make linking nicer in Empathy.

In the future, we'll be looking into making this process even easier.

folks-inspect

Our handy folks-inspect utility has gotten a bit of attention to support the fields added since 0.4.0, making it even more useful for debugging.

QtFolks

I've followed up on Marco's initial QtContacts bindings for Folks (QtFolks) by adding additional fields and supporting contact adds, edits, and removes. The git respository is here and release tarballs here (for now).

I hope we can get our KDE-using friends interested in Folks now that we've got some decent Qt bindings!

Demos

The moment everyone has been waiting for:


(In case your browser can't play WebM: download here; YouTube version here)

The demos in the video are done through the QtFolks demo program because Empathy filters out non-IM contacts (at least as of this writing). But we'll certainly be supporting all Folks contacts in the Gnome 3.2 contacts features.
hackergotchi

Folks 0.1.17 includes a migration tool

The latest release of Folks, 0.1.17, includes a new folks-import tool to import meta-contacts data from Pidgin. So if Pidgin knows frank.reynolds@example.com and the_penguin@example.org are the same person, folks-import will catch Empathy up-to-speed.

Now would be a good time to double-check your "Excuses for not switching to Empathy" list.

Thanks to Philip for writing the tool.

We've also got a nearly-working Pidgin backend to let Empathy support this functionality native in its Import wizard, but it was too late in the Gnome 2.32 release cycle to squeeze it in.

And farther into the future, Folks should perform reasonably aggressive auto-linking, which should make even semi-automated linking like this unnecessary.

--

If you've got AIM accounts, you may bump into bgo#629311. See the workaround.
hackergotchi

Empathy hacking setup

Edit: Good News, Everyone! Sjoerd pointed out that the D-Bus fiddling in this post is unnecessary (as I'd hoped, but hadn't been able to manage). I mistook bgo#628738 for misconfiguration on my system.

So, this "guide" can be shortened to:

  1. create a new user (empathy-dev)

  2. install jhbuild

  3. jhbuild build empathy

  4. echo "export PATH=/opt/gnome/sbin:/opt/gnome/bin:$PATH" >> /home/empathy-dev/.bashrc

  5. echo "export XDG_DATA_DIRS=/opt/gnome/share:/usr/share:$XDG_DATA_DIRS" >> /home/empathy-dev/.bashrc

  6. slogin -X empathy-dev@localhost

  7. empathy

In order to do development on Empathy, I've created another user on my system (creatively named "empathy-dev") and run my development versions of programs, libraries, and services through this user. This has a couple major benefits:

  • zero downtime - I frequently restart Empathy during development. Running the default system version of Empathy as my regular user means I can stay online as I develop.
  • fault isolation - if I horribly break my development environment (eg, the bleeding-edge libraries I depend upon introduce major bugs), it doesn't affect my system stability

It was somewhat non-trivial to create a good setup, so I thought other people might find this handy.

Throughout this short guide, I use references to /opt/gnome, which is the standard jhbuild $PREFIX. So substitute your development $PREFIX as necessary. I make a point to not use the standard Unix development $PREFIX (/usr/local) so that my regular user doesn't run the development programs or libraries (for stability and sanity; see below).


Skip the remaining steps this guide - it's needlessly complicated


First, we need to set up a few bits and pieces so your shell and D-Bus know where to find the development programs and config. I use a fairly standard D-Bus local session config file.

The important (and only non-standard, I think) part is:

/opt/gnome/share/dbus-1/services

This allows the programs installed in our $PREFIX to launch necessary services (eg, Mission Control, the Telepathy Connection Managers) through D-Bus activation.

Download this file, adjust it as necessary, and put it in /home/empathy-dev/.dbus-1

Add this to the end of your development user's .bashrc:

export PATH=/opt/gnome/sbin:/opt/gnome/bin:$PATH

# let programs find any new icons, default config files, etc.
export XDG_DATA_DIRS=/opt/gnome/share:/usr/share:$XDG_DATA_DIRS

# automatically start a D-Bus session bus
alias dbus-setup="dbus-launch --config-file=$HOME/.dbus-1/session-local.conf"
session_setup=$(dbus-setup)
if [ $? -eq 0 ]; then
echo "\nSet up D-Bus session for $USER:";
echo $session_setup
eval export $session_setup
else
echo "\nFAILED setting up D-Bus session for $USER";
fi

Once you've got that set up, it's easy to launch the development version of Empathy as a different user:

slogin -X empathy-dev@localhost

Then, just run empathy and marvel at its independence from your main user's instance. However, if you haven't installed a newer version of empathy in $PREFIX/bin, this will just fall back to your system-installed version.

Fix this by setting up jhbuild and running "jhbuild build empathy". Find something to do for about 20 minutes. When you come back, if our dependencies haven't failed to build, you'll be able to run the latest version of Empathy as empathy-dev, without getting in the way of your system user.

Happy hacking!

--
dbus-launch will keep the dbus-daemon (and anything it activates) running by default after your development user logs out. There's a --exit-with-session option, but for some reason, if I use it here, about 30% of my keystrokes at the shell don't arrive. I'm flabbergasted, but it seemed completely tied to this dbus-launch option (I tried several times each way).
hackergotchi

Meta-contacts in Empathy (and beyond!)

We've finally closed Gnome Bug #460647 (Empathy: Support metacontacts)!

Let's get right to the pictures:

metacontacts-linking-dialog

This is our "Link Contacts" dialog. Selecting a number of contacts on the left side designates which contacts will be included in the Individual (read: meta-contact) at right.

We currently auto-link contacts with the same service ID (foo@example.org) contained in multiple accounts. In the future, we're going to push as much into auto-linking as possible. My goal is that most users would never need to use this dialog. Until then, this dialog is how you designate meta-contacts in Empathy.

metacontacts-edit-dialog

If you accidentally link together contacts you didn't mean to, you can simply open the Individual's info dialog and hit Unlink to start over.

metacontacts-persona-menus

One of the first benefits of meta-contacts is pulling together multiple modes of contact for a person into a single place. So if you've got a friend with separate work and personal IM, you can be an upstanding citizen and only send funny cat pictures to their personal IM address. (If they sign on to both at work, at least you tried your best).

We've kept the 1-click Chat/Audio Call/Send File/etc. actions for the (I think, more-common) case that you just want to communicate and you don't care about the specifics. This just uses any valid (ie, online) account for that Individual.

Thanks

This meta-contacts functionality is all possible thanks to the Folks project. It's a collection of libraries (so far, all distributed together) that collect contact information through various backends (Telepathy so far; many more to come) and aggregate Individuals (ie, People). Empathy was our starting point, but we really plan to integrate very well into the Gnome desktop and beyond. Think: typing -Alice- to start a chat with Alice through the Gnome Shell.

I'll discuss Folks in greater detail in future posts.

I'd like to thank my colleagues Philip "code factory" Withnall for helping with Folks development/Empathy integration and Empathy maintainer Guillaume Desmottes for letting us know exactly what we broke in each commit. And, of course, a special thanks to our employer, Collabora, for sponsoring us to work on this full-time.

Try it now!

Since you're no doubt very excited about these new features in Empathy, you can use them Right Now™ in version 2.31.91 (also coming to Debian experimental shortly). Be sure to file any and all bugs you run into, so this will be as solid a stable release as any other!

If software bugs have you leaping onto the nearest elevated furniture, no problem -- Empathy will, of course, by included in Gnome 2.32, coming to a friendly Linux distribution near you!
hackergotchi

Empathy accounts work for MeeGo (and free desktops everywhere)

I'd like to congratulate the MeeGo team on shipping MeeGo 1.0 for Netbooks! It's nice to see user-experience-driven design on Linux (where it tends to be conspicuously absent), shipping in a complete product. It really shows, too -- the UI is extremely responsive and it makes the whole experience a joy.

Thanks to Intel hiring Collabora for some MeeGo work, I made a few contributions to the distro some months back. Some were evolution data server code clean-up, adding tests, etc. For the interesting, user-visible changes (that I didn't work on), see Michael Meeks's post about Evolution in MeeGo.

But the more visible contributions were to Empathy's account settings. Originally, Empathy's "first-run" account setup wizard and accounts settings dialog were in the same process as Empathy. This meant you had to launch Empathy before you could edit your accounts, which is a little awkward if you're another application. Telepathy accounts aren't Empathy-specific, so it makes sense that the dialog be independent (though it still lives in the Empathy source tree).

Empathy account wizard in MeeGo 1.0

Thomas Wood's unified Control Center work appears MeeGo, and I adapted the Empathy accounts dialog to embed in it. The screenshot is fairly self-explanatory. When you access the accounts from the Control Center, you get a nice, full-screen version of the dialog:

Empathy account wizard in MeeGo 1.0, embedded in the control panel

Michael Meeks mentioned how small most netbook screens are. When you also factor in the MeeGo panel at the top of the screen, you can get situations where you need scrolling (but didn't on the full-sized desktop). So we added a scrollable view to the accounts dialog (to fit the full content of the Advanced section for protocols with a bevy of bells and whistles):

Empathy account wizard in MeeGo 1.0, embedded in the control panel

Since the actual dialog content is the same whether it's embedded or not, the same functionality works when it's stand-alone as well:

Empathy account wizard in MeeGo 1.0, advanced settings

And because the MeeGo team understands open source software, these features were developed out in the open and have been included in the last few releases of Empathy. So if you don't have them already, you will soon!
hackergotchi

Google frees VP8, announces WebM (with a little help from Collabora Multimedia)

This morning I was thrilled to hear that Google has opened up their high-performance video codec, VP8 and announced WebM, a container format with VP8 video and Vorbis audio. So now we've got a media format that's high-performance (not Flash), has low-power-requirements (not Flash) and is fully open and freely usable (not Flash, not H.264). The only missing piece is making sure it's widely supported (like Flash and, to a slightly lesser extent, H.264). So if this takes off, we can expect a much better experience for audio, video, animation on the web and on our phones.

There are a number of important players supporting this effort, including Collabora (GStreamer and our projects that use GStreamer, like Telepathy for multimedia calls), Google (YouTube, Android, Chrome, Chrome OS), Mozilla, Opera, Skype, Adobe (through Flash, surprisingly) in terms of software. And practically every major hardware player as well (AMD/ATI, ARM, Broadcom, Freescale, Logitech, nVidia, Qualcomm, TI, and more). This should make the problem of hardware support (which has traditionally been a problem for open media formats) a moot point.

Collabora Multimedia and Entropy Wave have added WebM/VP8 support to GStreamer, so support is already available for the open desktop.

Exciting times!
hackergotchi

LCA 2010 - Day 3

Day 1 | Day 2

Andrew Tridgell - FOSS and Patents

This informative talk went into a bit of detail on the current situation with (software) patents and some best practices for open source projects who want to avoid litigation. First, if you do get contacted by a company who claims your project violates one of their patents, contact public defenders, such as the Software Freedom Law Center or Electronic Frontier Foundation. In general, it's useful to know how to read patents, such as reading the abstract (but not stopping there, since they can often be misleading), then skipping ahead to the individual claims (which are the core of the patent), and then referring back to the diagrams, definitions, etc. as necessary.

And important point is that the "prior art" defense that so many people like to cite is actually very hard to win. This requires that the prior art covers every single claim in the patent. Instead, a "non-infringement" defense only requires that your work not (exactly) match every independent claim in the patent. There tend to be far more dependent claims than independent claims, so this defense tends to be easier to win.

An interesting point Andrew brought up was that he doesn't think it's to our advantage to avoid reading patents. His point is that even single damages (in the case that you unknowingly infringe a patent) is enough to end an open source project, so if you are forced to pay triple damages (in case you are aware of the patent), that doesn't change the end result. I unfortunately didn't have time to ask him this: wouldn't it matter to the author themselves whether they have to pay $LARGE_SUM_OF_MONEY or 3*$LARGE_SUM_OF_MONEY out of their own pocket? In either case, it's an ugly situation that you can be punished for being more knowledgeable.

Finally, Andrew suggested this strategy for open source projects to make themselves worse targets for patent suits (which is really all that matters): find and widely publicize work-arounds to patents. Closed-source companies are unlikely to do so, since a work-around could be a business advantage over another competitor who might otherwise be forced to license the patent. That way, we'll appear (and effectively be) a lot more work for patent owners to troll. That way, we can build up a reputation for not being worth the hassle. A great thing to aspire to.

20100123_007.jpg

Paul Mackerras -- Perf Events (in the kernel)

A replacement for perf counters, perf events provide kernel probes and simple API to perform fine-grained performance benchmarking on a Linux system. Whenever possible, perf events use hardware to get the most-precise data possible. There's a reasonable software-only fallback based on the high-resolution system clock (which most systems support at this point).

The API is essentially just one system call (which returns a file descriptor) and specific content behavior from that virtual file. Everything else is read(), close(), etc.

Events can be per task, per CPU, and, recently added, per-task-per-CPU. Per-task tracking can be recursive (forked processes get a copy of the parent counter struct and its final values are added to the parent upon exiting or explicitly synchronizing).

Perf events can trace cache (including TLB) activity, page faults, context switches, CPU migrations, and data alignment and instruction emulation traps.

Useful benchmarking always starts with good tools, and it sounds like we're finally getting a great tool with perf events!

Rusty Russell -- FOSS Fun with a Wiimote

Rusty detailed his geeky plans for making sure his daughter grows up to be a geek. These mainly involve writing some software to (ideally) translate her hand movements to actions on their TV. The idea was to help her establish causality at an early age, I think.

Anyway, I can't really do it justice in writing. You'll have to wait until the video is posted to see for yourself.

20100118_107.jpg

Carl Worth -- Making the GPU do its job

After giving a brief history of computer graphics an GPU development, Carl explained how hardware graphics support has sort of oscillated between discrete and integrated (into the CPU) hardware. (I wasn't aware that we'd already made the integrated → discrete → integrated cycle once before our current (in some contexts) migration back to integrated graphics).

The main problem with graphics in Linux right now is that we often have bad performance (which also means bad power consumption) and we need two drivers per video card (family, at least) -- one for 2D and one for 3D.

In order to figure out our bottlenecks, Carl created cairo-trace to measure the actual performance. This nifty program records the timing of cairo API calls for any program running under it. These traces can be played back through cairo at any time (at maximum speed), to continually improve performance for real-world uses of Cairo. I'm not sure if it's already in place, but these could easily be added to the (from what I hear, very good) automated cairo test suite, to avoid releasing regressions. If only more open source projects took testing and performance this seriously!

As it turns out, the current performance in most cases was actually better in pure software than in cairo-xlib (the latest stable backend for cairo).

An experimental GPU-accelerated backend (cairo-drm), in which Cairo bypasses X and uses GEM to render directly in the kernel, improves performance dramatically (10x speedup for Firefox). But the caveat is that it requires yet another driver per video card (bringing us to 3 total, for those of you keeping track at home).

Another approach, cairo-gl, which has Cairo bypass X to render to MESA, requires only 2 drivers total and should eventually have performance closer to that of cairo-drm. But for the moment, its performance is much worse than cairo-xlib.

Robert O'Callahan -- Open video on the web

Video on the web right now has two major players: Flash and Silverlight. (Maybe I just took ambiguous notes, but it's obvious that Silverlight is nowhere near Flash in terms of marketshare).

Beyond the obvious problem of software freedom, Flash is notoriously unstable (apparently a huge percentage of crashes of OS X are directly Flash's fault). Mozilla has decided that it's time to do something to push open media formats, to ensure this important (and growing) chunk of the Web retains the openness that has made the rest of it so popular and useful.

Some questions about the open formats:
  • Should we ignore patents?
  • This is only feasible while open formats are irrelevant, which isn't a great strategy.
  • Wait for the patents to expire?
  • They'll just be replaced with the next closed format.
  • Just pay the licensing fees?
  • Using a codec means you need to license it for your market size and per viewing (according to MPEGLA's fee structure), which is impossible for most websites serving content and for Mozilla to pay for distributing Firefox.

Another issue: video is not just about YouTube (passive watching). Flash adds interactivity (related videos, captioning, relevant ads, etc.), so we'll need an open counterpart to this as well.

Mozilla's solution for open video is Ogg Theora, which has had nice advances lately. GStreamer developer David Schleef got decent performance on OMAP3's DSP (in the N900 and other embedded devices) for Theora. The Ogg Index project adds indexing to the Ogg (container format?) to fix stream seeking over the web (which is frequently unusable without an index).

Mozilla has been shipping Theora support in Firefox 3.5+, since it doesn't want to pay, can't pay, and shouldn't pay for licensing codecs. If suddenly software patents were invalid, it'd be fine to just standardize on h.264, but that probably won't happen.

A big part of the chicken-and-egg push for open formats requires working with content providers and distribution networks. So Mozilla has gotten Dailymotion, the Internet Archive, and other websites to support Ogg Vorbis and Theora.

The other half is getting the browser to support the formats. Firefox 3.5+ can handle a Theora <video> tag; Chrome ships Theora (and H.264) support; Opera will ship Theora only. In FireFox 3.6, we'll get fullscreen Theora. Also in the pipeline is GPU-accelerated playback and Mobile/Maemo optimizations.

Partial successes in this push to open video include Vimeo and YouTube planning to support HTML 5's <video> tag on their sites (though they'll only be using the non-free H.264 codec).

So open media formats are slowly advancing on the web. We're basically at the point where most web standards are friendly to free software; if Gecko and (or) WebKit can't implement it, people don't propose it as a standard.

20100121_003.jpg