Serendipitous altruism

The quality of OpenStreetMap‘s work speaks for itself, but it seems that we need to speak about it too – especially now that Google is attempting to to appear as holding the moral high ground by using terms such as “citizen cartographer” that they rob of its meaning by conveniently forgetting to mention the license under which the contributed data is held. But in the eye of the public, the $50000 UNICEF donation to the  home country of the winner of the Map Maker Global Challenge lets them appear as charitable citizens.

We need to explain why it is a fraud, so that motivated aspiring cartographers are not tempted to give away their souls for free. I could understand that they sell it, but giving it to Google for free is a bit too much – we must tell them. I’m pretty sure that good geographic data available to anyone for free will do more for the least developed communities than a 50k USD grant.

Take Map Kibera for example :

“Kibera in Nairobi, Kenya, widely known as Africa’s largest slum, remains a blank spot on the map. Without basic knowledge of the geography and resources of Kibera it is impossible to have an informed discussion on how to improve the lives of residents. This November, young Kiberans create the first public digital map of their own community”.

And they did it with OpenStreetMap. To the million of people living in this former terra incognita with no chance of profiting a major mapping provider, how much do you think having at last a platform for services that require geographical information without having to pay Google or remain within the limits of the uses permitted by its license is worth ?

I answered this piece at ReadWriteWeb and I suggest that you keep an eye for opportunities to answer this sort of propaganda against libre mapping.

Today I mentioned that 15 years late, I had finally put a name on a past adolescent problem : patellofemoral pain syndrome (PFPS). As far as I understood, it is a growth related muscle unbalance that solves itself when the body reaches maturity.

As usual with most of my microblogging, I dispatch the 140 chars to several sites using Ping.fm and then follow the conversation wherever it eventually happens. In that case, a conversation developed on Facebook. Friends asked questions and gave their two cents – business as usual.

And then an interloper cut in : “Jean-Marc we can help correct your patellfemoral pain syndrome. It is the mal-tracking of your patella. Check us out at mycommercialkneesite.com”. It is not entirely spam at first sight because it is actually on-topic and even slightly informative. But it is not really taking part in the conversation either because it is a blatant plug for an infomercial site. So spam it is, but cleverly targeted at a niche audience.

I does looks like all the blatant plugs that we have been seeing for decades in forums and mailing list – usually for a short time after which the culprit mends is devious ways or ends up banned. But there is an innovative twist brought by the rise of the “real-time web” : the power of keyword filtering applied to the whole microblogging world is the enabler of large-scale conversational marketing. Obnoxious marketers attempting to pass as bona fide contributors to the conversation are no longer a merely local nuisance – they are now reaching us at a global scale and in near real-time.

Marketers barging in whenever someone utters a word that qualifies their niche are gatecrashers and will be treated as such. But I find fascinating that we now have  personalized advertising capable of targeting a niche audience in real-time as the qualifying keywords appear. Not that I like it, but you have to recognize it as a new step in the memetic arms race between advertisers and audience.

Imagine that coupled with voice recognition and some IVR scripting. Do you remember those telephone services where you get free airtime if you listen for advertising breaks ? Imagine the same concept where during the conversation someone – a human, or even a conversational automaton – comes in and says “Hey, you were telling your boyfriend about your headache ? Why don’t you try Schrufanol ? Mention SHMURZ and get the third one free !”.

Even better, add some more intelligent pattern recognition to go beyond keywords. The hopeless student who just told his pal on Schmoogle FreeVoice telling about his fear of failure at exams will immediately receive through Schmoogle AdVoice a special offer for cram school from a salesdrone who knows his name and just checked out his Facebook profile. You think this is the future ? This is probably already happening.

The October 2008 article “American troops in Afghanistan through the eyes of a French OMLT infantryman” gathered more than two hundred comments and will be past a hundred thousand visits by most reasonable accounts before the year ends (see 2008 traffic and 2009 traffic). I though that translating this piece would surely raise some interest, but I never expected it to be that much. More than one year later it is still sparking interest among citizens of the United States. During the Bush era, the image of France among the right wing in the United States seems to have suffered a lot, and as a result a lot of people have been genuinely surprised to read an article that showed that in spite of the politics we actually manage to work together with cordial relationships.

One year later, I am still receiving mail asking about the source of the article, from readers who enquire about its authenticity. Considering the unofficial reactions from members of the French armed forces and from readers with interests in the defense community, I have a rather good certitude about the authenticity of the original essay. With the source blog defunct and having lost touch with the original author who was not seeking public exposure and only made a couple of fleeting comments before disappearing from the media landscape, I am unable to prove anything. The author went by the pseudonym of “Merlin” and his blog was called “Le Blog de Merlin” at http://omlt3-kdk3.over-blog.com. The disappearance of the original article’s page is also a pity because that is where I exchanged comments with the author.

I do not believe that the author ever thought that his blog would get noticed significantly, even in France. It was featured in a well known blog by Jean-Dominique Merchet, a military journalist at the French daily “Liberation” who has an excellent reputation for reliability – it is his post that got the ball rolling. The original author probably does not even realize now how many American blogs and forums have been discussing his article.

Since then I have lost track of him and I do not know his real name : though military blogging is rather common in the United States, it is still quite alien to the more conservative culture of the French defense community, so it seems that most military-related people in France prefer the pseudonymous discussions in forums to the more public exposure of blogs and even there they won’t take much risks in expressing themselves – the French army is not nicknamed “la grande muette” (“the great silent”) for nothing. This rarity may be one of the reasons why this humble piece of French first hand account of recent events won such attention. But no one here expected this – to us cheese-eating surrender monkeys, the United States of America are always full of surprises !

Of course, as it benefits our relationship and the image of France in the United States, the original article and even my translation could have been an elaborate psyop by the French government. Or it could be the work of shadowy pro-French non-governmental propaganda outfits. Or a fake by someone who wants everyone to believe in one of those two hypothesis in order to later appear as exposing the evil scheming French. At that point, we enter the realm of conspiracy theories and I’m sure that some will have a great time speculating about it. But from where I sit, there is coherent case for this story to be just what it appears to be : a simple account of good working relationships.

As Ars Technica announced three days ago,  Intel’s 2009 launch of its ambitious Larrabee GPU’s has been canceled : “The project has suffered a final delay that proved fatal to its graphics ambitions, so Intel will put the hardware out as a development platform for graphics and high-performance computing. But Intel’s plans to make a GPU aren’t dead; they’ve just been reset, with more news to come next year“.

I can’t wait for more news about that radical new architecture from the only major graphics hardware vendor that has a long history of producing or commissioning open source drivers for its graphics chips.

But what are we excited about ? In a nutshell : automatic vectorization for parallel execution of any known code graph with no data dependencies between iterations is why Larabee is about. That means that in many cases, the developper can take his existing code and get easy parallel execution for free.

Since I’m an utter layman in the field of processor architecture, I’ll let you read the word of Tim Sweeney of Epic Games, who provided a great deal of input into the design of LRBni. He sums up the big picture a little more eloquently and I found him cited in Michael Abrash’s April 2009 article in Dr. Dobb’s – “A First Look at the Larrabee New Instructions” :

Larrabee enables GPU-class performance on a fully general x86 CPU; most importantly, it does so in a way that is useful for a broad spectrum of applications and that is easy for developers to use. The key is that Larrabee instructions are “vector-complete.”

More precisely: Any loop written in a traditional programming language can be vectorized, to execute 16 iterations of the loop in parallel on Larrabee vector units, provided the loop body meets the following criteria:

• Its call graph is statically known.
• There are no data dependencies between iterations.

Shading languages like HLSL are constrained so developers can only write code meeting those criteria, guaranteeing a GPU can always shade multiple pixels in parallel. But vectorization is a much more general technology, applicable to any such loops written in any language.

This works on Larrabee because every traditional programming element — arithmetic, loops, function calls, memory reads, memory writes — has a corresponding translation to Larrabee vector instructions running it on 16 data elements simultaneously. You have: integer and floating point vector arithmetic; scatter/gather for vectorized memory operations; and comparison, masking, and merging instructions for conditionals.

This wasn’t the case with MMX, SSE and Altivec. They supported vector arithmetic, but could only read and write data from contiguous locations in memory, rather than random-access as Larrabee. So SSE was only useful for operations on data that was naturally vector-like: RGBA colors, XYZW coordinates in 3D graphics, and so on. The Larrabee instructions are suitable for vectorizing any code meeting the conditions above, even when the code was not written to operate on vector-like quantities. It can benefit every type of application!

A vital component of this is Intel’s vectorizing C++ compiler. Developers hate having to write assembly language code, and even dislike writing C++ code using SSE intrinsics, because the programming style is awkward and time-consuming. Few developers can dedicate resources to doing that, whereas Larrabee is easy; the vectorization process can be made automatic and compatible with existing code.

With cores proliferating on an more CPUs every day and an embarrassing number of applications not taking advantage of it, bringing easy parallel execution to the masses means a lot. That’s why I’m eager to see what Intel has in store for the future of Larrabee.