Weekly Head Voices #127: Readers are leaders.

Betty’s Bay Beach impression by Genetic Offspring Unit #1, age 11.
  • This week I worked on automated email analysis and storage for side-project #38465 (more on this in future editions) and on bits of UI for a wxPython desktop app (yes desktop app! some of us fortunately still get to make them!) for my current main work project.
  • Had to make screencast to demonstrate milestone deliverable of above-mentioned main project. Making screencasts is an obscure but longstanding hobby of mine, but I needed to level up slightly, so the business bought me ScreenFlow 7.2. For the first time ever, I recorded the screencast in multiple segments and did the voice-over later. Soon these new skillz will trickle down to my publically available screencasts.
  • On that topic, having a good microphone is crucial, not only for screen recordings but also for video meetings. I recently acquired the Samson Go Mic to complement my larger Samson C01U. The Go is brilliant: Recorded voice quality comes close to the C01U in spite of the Go’s compact form factor, and it has a hardware switch to select either of the the built-in omni-directional, for meetings, or cardioid, for more dedicated voice recording, microphone elements.
  • Ironically, an ex-colleague posted “How to Fix Facebook—Before It Fixes Us” on Facebook, a long and worthwhile read on how FB is used to spread fake news that effectively manipulates public opinion, and what should be done to remedy this. Here is a choice quote to get you started:

We still don’t know the exact degree of collusion between the Russians and the Trump campaign. But the debate over collusion, while important, risks missing what should be an obvious point: Facebook, Google, Twitter, and other platforms were manipulated by the Russians to shift outcomes in Brexit and the U.S. presidential election, and unless major changes are made, they will be manipulated again. Next time, there is no telling who the manipulators will be.

  • In the same vein, I continuously try to spend as few as possible minutes on YouTube, but the one thing I will definitely continue watching is Károly Zsolnai-Fehér’s brilliant Two Minute Papers channel! Most recently, his treatment of Distilling a Neural Network Into a Soft Decision Tree, a paper by Nicholas Frost and Geoffrey Hinton, caught my interest. In this, they address the problem of neural network explicability (it’s hard saying at a higher level why a neural network makes a particular decision) by deriving a soft decision tree from that trained neural network. The tree is not as accurate as the network, but is able to give plausible explanations for the network’s decisions. See the 4 minute long two minute paper video (hehe) here:
  • I came across the following on reddit, again quite ironically, and I have since taken to saying it to my genetic offspring units (GOUs) at every possible opportunity:

Readers are leaders!

Have a great week readers, I hope to see you again really soon!

Weekly Head Voices #15: Auto-tune my cloud.

In this 15th edition of my weekly head voices, I move yet more of my life into the cloud, discover (years after everyone else) the delightful auto-tune internet meme and finally go all backyard-psychological whilst staring into the distance, obviously defocused, and waxing on about the purpose of this weblog.

Before continuing, you might like to watch this clip explaining why you shall build a turtle fence (I’ll get back to the clip after my dropbox story):

Last week I completed, you guessed it, 21 GTD tasks spread over 10 projects. Once again, one more task than last week. The question is thus not if, but when I’m going to have to disappoint you. :)  Worth mentioning on the miscellany front is that I’ve started playing around with processing, a fantastic little system for programming visual effects and interaction, in preparation for a new first year course. My goal is to get the students irrevocably addicted to the coolness that is media processing! I’ll keep you up to date…

In a previous post, I was quite enthusiastic about Dropbox and its possibilities for collaboration. As some of you might now, I really like this whole living-in-the-cloud idea: I use GMail, Google Calendar and Google Documents quite extensively and I’m even paying for extra storage with the big G.  So, during the past week, I decided to bite the bullet some more and to move 12G more of my data right into the cloud, courtesy of a 50G Dropbox Pro account.  Up to now, I had a ridiculously complex synchronisation system keeping various subsets of my data up to date between a netbook, a laptop and three different servers. At the core of this system was unison, a brilliant multi-way open source synchronisation tool. In spite of this system mostly working, its complexity and the starkly contrasting It-Just-Works nature of Dropbox convinced me to give the simple solution a shot.

So far I can only report that I remain impressed: At one stage I manually copied a complete dropbox (12Gigs) from one already synced machine to a fresh target machine and started the Dropbox software on the target. It politely asked:

There is already a folder in your home folder called Dropbox. Do you want to merge all the existing files in that folder into your dropbox?

After clicking on the “HELL YEAH!” button (that’s how it felt, ok), the software went on indexing for a minute or two and then correctly claimed that everything was nicely synced up. Very much understated robustness, kudos to the developers. I’m going to test-drive this whole business for one month, and then let you know whether it’s going to be  a permanent fixture in my cloud-home.

Still wondering why you should build a turtle fence? Well, you can blame the Auto-Tune internet meme. Very shortly, auto-tune is an audio effect that corrects one voice to be perfectly in tune with backing music. In other words, a vocalist who can’t sing is in fact no problem at all, computer will fix! Initially it was used quite sparingly and its application was even sometimes kept a secret, until artists such as Cher and especially T-Pain turned it into an art form, in fact exaggerating the effect until it gave a decidedly unsubtle robotic voice effect. The effect has become so famous that it now gets to call itself an internet meme and is often parodied.  The turtle clip above is just one of a whole series (auto-tune the news, see them all!). In the clip below, internet scientists *ahem*, including the well-known Professor Weird Al Yankovic, take an in-depth look at this phenomenon:

Finally, back to the purpose of this weblog… Good blogs all seem to have some central theme, such as photography, environmental issues, science or pokemon. I seem to recall that I’d also seen this in more than one “how to become an A-list blogger” guides.  I don’t find it hard to believe that this is very important.  However, this blog has never had a central theme, it’s always been me blabbing about the various things that I find blab-worthy.  I’ve never been able to come up with something better, and it was definitely not for lack of trying.  The Weekly Head Voices, by focusing my blabbing into slightly more coherent episodes, have finally helped me to come to a conclusion.  Besides acting as a creative outlet, sitting down every week and carefully externalising a specific subset of my experiences with the express purpose of having it read by a small number of people, is an important ritual during which I am forced to distance myself from the events of the week, and to self-reflect.  By formally concluding the previous period in this way, one has the mental room to manoeuvre in preparation for the next. If you by any chance find any of it entertaining, or at least you just can’t look away, it’s a win-win situation!

In other words: Theme-schmeme! The voices in my head will continue to be the many and various topics of this weblog, thank you very much. :)

Have a great week kids!

Weekly Head Voices #7: The Answer, a STAR is born, post-human YouTube

I know that I said that I’d excuse myself from writing one or two editions of the Weekly Head Voices due to the upcoming VisWeek 2009 (and 2010, yes I will be blogging from the future) and the live-blogging explosion it will most likely lead to, but I somehow couldn’t stay away from my beloved WordPress installation. Oh well, so be it!

Welcome to the Weekly Head Voices #7, a lucky number for a blog post dealing with week #42, a week that could potentially have something to do with the answer to life, the universe and everything. Auspicious is an understatement.  (Oh my word, I just realised that week #42 is only starting as I write this.  Week #41 is the one just behind us and hence the subject of this post.  Oh well, just add 1 for the answer to life, the universe and everything!)

First the customary (but more uplifting than usual) visual element, I’ll get back to it at the end of this post.  Watch it to get in the mood:

A brilliant and excellently debauched Tuesday evening with friends in Rotterdam led to an exceedingly difficult Wednesday, the difficulty of which was definitely not alleviated by my 6 scheduled meetings. Fortunately a number of the meetings turned out being quite energising, leaving me with an energy surplus by the end of the day.  Beer: 0 Charl: 1.

On Thursday and Friday, Jorik and I helped the famous Jelle Feringa with a workshop he set up to expose architecture M.Sc. students to the Wonderful World of Scientific Visualisation (!!!). Jelle has somehow managed to bring our two fields together with great effect, and was trying to impress some of this magic onto the students. I found it quite challenging explaining bits of Python and VTK programming to a class of students that had for the largest part never seen source code before, but by the time we had gotten to extracting surfaces and direct volume rendering with DeVIDE, things were going just swimmingly. At least that was my impression. :)

On Friday, I submitted a rather extensive proposal to write a Eurographics State-of-the-art Report (STAR) together with a number of similarly interested colleagues. I’d love to tell you what the STAR’s about, but would prefer waiting until we know whether it has been accepted or not. In any case, if it is indeed accepted, we will have the privilege of publishing the full-length survey article in the Proceedings of the Eurographics, giving a 90 minute (!!!) presentation during the conference next year, and submitting a revised version of the article to the journal Computer Graphics Forum. That’s like Graphics/Visualisation-scientist crack man!

Because this is my little soap-box, I have decided to conclude with one of my typical “I love the 21st century” RA RA moments.

First, as you all know, one can easily upload photos and videos to facebook by mailing them to a special per-user email address: Go to facebook.com/mobile – the special email address is on the bottom left under “Upload Photos via Email”. This is especially useful for sending absolutely useless but seriously incriminating media directly from your cellphone, especially exactly when you shouldn’t be doing so, for example at 3 in the morning.

I’m probably one of the last to come across this, but youtube of course has similar functionality! Login, then go to your account page, then settings and finally mobile setup to find your upload email address. This means you can upload your incriminating media at 3 in the morning to yet another site, except that this one has a far, far greater reach than facebook.

Is this why I love the 21st century? Well, it’s a minor reason amongst many others. A slightly more meaningful one is the fantastically uplifting video clip embedded at the start of this post. That’s a real wedding (and what a wedding it was!), the clip was posted in July of this year and has since been watched more than 27 million times. That’s crazy! In what other century can a visual memory of your humble, although very original, wedding be seen 27 million times and so brighten up the respective days of millions of people that won’t ever even know you?

That’s just so beautiful, it’s post-human.

P.S. The sound-track to the clip is Forever by Chris Brown. Instead of suing the love-birds, Sony (the label) and Brown decided to work together with Google to monetise the occasion. Notice that when playing the video, there’s a box underneath with the name of the song and a “Buy Song” button. It turns out that due to this video, the single was slung back into the top 10 charts, more than a year after its release. This GoogleBlog post describes how everybody involved wins. It’s gratifying that labels are slowly waking up to the new order. May those that refuse to smell the coffee wither away as quickly as their irrelevance seems to indicate.

P.P.S. My mother now has a Blackberry.

Recording screencasts on Windows with free software

(This post was first written in May of 2008, but I’ve been updating it periodically. See also the updates right at the end.)

Screencasts refer to video recordings of screen activity, often with narration. These can be used to demonstrate software or to serve as a kind of visual HOWTO. We often make screencasts of software we design in the Medical Visualisation group at the TU Delft to use in presentations at conferences or to distribute online.

On Windows, Camtasia Studio ($300) or Camtasia SnagIt ($40) are probably the best options your money can buy. Most of the free alternatives suck quite badly: This includes the Windows Media Encoder, thank you very much. In fact, the Windows Media 9 Screen Capture Codec has been fine-tuned to create the worst possible quality movies you can imagine. Another problem with the free options is that they often can’t sustain writing the video stream to disk, hence resulting in dropped frames and unusable screencasts. When they are able to sustain writing, it probably means that the compression is completely killing video quality.

Fortunately, it turns out that there is a free option which offers comparable performance to the Camtasia products, and for good reason. It’s called CamStudio, and it’s even open source! It’s terribly important that you also install the lossless CamStudio Screen Capture codec, it’s this that makes all the difference. This codec compresses all frames with the fast LZO lossless compression algorithm, so you get the highest possible quality without dropping frames due relatively slow disc writing.

Using CamStudio, I made a 3.5 minute screencast, with live audio recording, show-casing some of the new DICOM browsing functionality in the next DeVIDE release. After capture, I transcoded the CamStudio screen capture codec AVI to XVID using MediaCoder, and then uploaded to YouTube (play at your own risk!):

Update on 2010-02-21:

I’ve made two more screencasts and learnt some time-expensive lessons:

  • DON’T encode with H.264, in spite of YouTube’s recommendation. This screws up captured text elements extremely badly. I had much more success with XVID at 1500 kbit/s and MP3 audio using MediaCoder, and THE MOST SUCCESS OF ALL just directly uploading the lossless codec screencast to YouTube, without any transcoding.
  • DO capture at 640×480 or at 1280×720 (HD).  For example 800×600 is NOT worth it, it gets downscaled by YouTube. By capturing at 640×480 or 1280×720 you have much more control over what finally appears on YouTube.
  • If you get audio / video sync problems (I experienced them with a 10 minute screencast), try activating “Use MCI Recording” in the CamStudio “Audio Option for Microphone”.
Update on 2011-10-22:

With CamStudio 2.6 r294 getting the codec working on Windows 7 64bit is not straight-forward. You first install the codec in the normal way (with the exe installer), then you have to do TWO things. First copy camcodec.dll from Windows\System32 to Windows\SysWOW64. Second, create a reg file with the following contents:

Windows Registry Editor Version 5.00
[HKEY_LOCAL_MACHINE\SOFTWARE\Wow6432Node\Microsoft\Windows NT\CurrentVersion\drivers.desc]
"camcodec.dll"="CamStudio lossless codec [CSCD]"
[HKEY_LOCAL_MACHINE\SOFTWARE\Wow6432Node\Microsoft\Windows NT\CurrentVersion\Drivers32]

Install it by double-clicking the reg file, or doing right-click and then selecting Merge. After this, you should be able to select the camstudio codec from Options | Video Options | Compressor.

Furthermore, you can upload camstudio-encoded movies directly to Google Picasa, it Just Works(tm)!

Update on 2013-02-13:
If CamStudio 2.7 complains about MSVCR100.dll being missing, you should install the MS Visual C++ redistributable (32bit, because CamStudio is 32bit).