[Thu Jan 31 11:55:39 CET 2008]

The New York Times published a couple of days ago a very interesting article on evolution written by Olivia Judson where she argues that evolution, far from being as random and chaotic as many people think, repeats itself very often:

Here's an evolutionist's dream: 10,000 planet Earths, starting from the same point at the same time, and left to their own devices for four and a half billion years. What would happen? Could you go on safari from one planet to the next seeing an endless procession of wildly different organisms? Or would many of the planets be home to life forms that are broadly similar?

The conventional answer to this question —the one championed by the late Stephen Jay Gould, for example— is that chance events, from mutations to asteroids, play such a large role in evolution that each of the planets would be totally different. And probably, after four and a half billion years, they would be. I wish we could do the experiment, though. It might hold some surprises.

Looking around the Earth, it's striking how often similar traits evolve in similar environments. So: birds living on remote islands typically lose the power of flight. Males in species (be they chimpanzees or yellow dung flies) where females are promiscuous tend to evolve high sperm counts and large testes. Animals that live in caves lose their eyes and their color: whether they live in Rwanda or Romania, they're a pallid, blind lot, the troglodytes. Mammals that specialize on eating leaves —be they cows or langurs (that's a monkey)— have evolved foreguts where bacteria break down the leaves, as well as special enzymes to help with digestion. Amazingly, the same phenomena are also seen in the hoatzin, a leaf-eating bird from South America. In short, evolution has a remarkable tendency to repeat itself.

So, what's the answer to the pluzzle? The only one that occurs to me is, obviously, that although evolution does work through random events in the short term, things are quite different when we view it from afar. Precisely because we are dealing with a gradual process, there are too many variables conditioning the outcome to truly allow a high level of randomness. After all, evolution has to work within the constraints that are given to it. There is no such a thing as a Deus ex machina. {link to this story}

[Wed Jan 30 14:38:53 CET 2008]

I just stumbled upon the arch-famous Think Different commercial from Apple:

{link to this story}

[Tue Jan 29 14:05:29 CET 2008]

Once again, science comes to prove common sense wrong. It's not the first time it happens. Actually, it happens all the time. Personally, I think common sense is overrated. Most people use it instead of the word "sensible", which is what they mean most of the time. Common sense tends to equal whatever has always been widely accepted (i.e., what we inherited from our ancestors), which is not necessarily true nor right. If a lie repeated a thousand times doesn't become truth, a lie passed down the generations doesn't either. Yet, we see thousands of them. Today, I read that a group of South-African scientists have discovered that chameleons truly don't adopt the color of their surroundings to seek protection against their predators but rather as a way to communicate with (and impress) each other, at least in the variesties of the species they used for their tests:

The authors suggest that limited color change might have evolved originally to function in thermoregulation or camouflage but that the dramatic changes in hue shown by some chameleon species evolved subsequently to enable chameleons to communicate using bright, flashy colors. These findings have broader implications for the evolution of animal social signals, since they support the importance of selection for signals that are highly detectable within the animal's environemtn and demonstrate how studies that take animal visual systems into account can be used to understand the evolution of signal diversity in animals.

Among other things, they found out that changing their colors to camouflage in their surroundings wouldn't help chameleons much, since their main predators would still perceive them anyways because their visual system was not so dependent on color. In other words, this could have been a classical case of us (humans) extrapolating our way to see things to the world out there and reaching the wrong conclusions. {link to this story}

[Tue Jan 29 13:41:46 CET 2008]

CIO magazine published an interesting piece on when (and when not to) use Perl for a given project. Nothing earth-shattering, but sort of interesting nonetheless: use it for pattern matching, in-place editing, replacement for shell scripts when the syntax gets too hard to follow, as a database manipulation tool or a cross-platform development language: but do not use to write real-time or high-performance applications (doh!), as a replacement for every shell script (simply put: there are times when it makes sense, and some others when it doesn't), as a web scripting language (the author argues that PHP or Ruby on Rails are far more maintainable) or in an obfuscated fashion. As it tends to be the case, make sure you use the right tool for the job at hand. Just keep that in mind, and you should be safe. There are some statements from the article I could nitpick on, though. For instance, why not recommend Perl for web development? The author argues that more modern languages offer "a cleaner integration into the webpage experience", and recommends to avoid Perl for "traditional CGI-style form processing" because "this code tends to be hard to read and maintain because the HTML ends up inlined inside the Perl code" but I'm not sure the picture is much different in the case of PHP, where you end up with something that could be seen as PHP inlined inside HTML or the other way around... it's just not clear enough which. {link to this story}

[Fri Jan 25 16:32:55 CET 2008]

While checking out the latest entries at The Edge (excellent website, by the way), I came across a few excerpts from a recent discussion between Richard Dawkins and Craig Venter at the Digital Life Design (DLD) conference in Germany that drew my attention. I found the following words from Richard Dawkins especially suggesting:

It's more than just saying you can pick up a chromosome and put it in somewhere else. It is pure information. You could put it into a printed book. You could sent it over the Internet. You could store it on a magnetic disk for a thousand years, and then in a thousand years' time, with the technology that they'll have then, it would be possible to reconstruct whatever living organism was here now. What has happened is that genetics has become a branch of information technology. It is pure information; it's digital information; it's precisely the kind of information that can be translated digit-for-digit, byte-for-byte into any other kind of information.

It's an idea I have been toying with lately. The more I read about the latest trends in both the natural and social sciences, the more I realize things appear to be converging towards a new paradigm centered around the concepts of information and network models. {link to this story}

[Thu Jan 24 13:46:59 CET 2008]

Information Week has published an interesting piece on how to roll your own Linux distro, precisely now that I was considering the possibility of building a customized version of Ubuntu for some non-profits I volunteer for. Most of the article is about Puppy Linux and Linux Fron Scratch but it's a good read, nonetheless. {link to this story}

[Tue Jan 22 15:16:21 CET 2008]

MIT Technology Review has published an interesting article on instant boot-up. In reality, it's a short piece about Slashtop, an OS based on Linux that's implemented on the BIOS itself and allows the user to access the Net almost instantly. This innovative concept makes it possible to browse the Web, chat or send emails just seconds after turning on the system, as long as the motherboard includes Slashtop. For the time being, only Asus has shown interest. {link to this story}

[Tue Jan 22 10:43:41 CET 2008]

You know things are bad when a reputed magazine publishes an article explaining how to downgrade from Windows Vista to XP and then decide to search Google for the terms "downgrading from vista to xp" and it returns a total of about 15,600 results. The intro to the article published on Information Week says it all:

If sluggish performance and numerous bugs in your shiny new PC running Windows Vista have got you down, you can downgrade —at no additional cost— to Windows XP. It's not hard. Here's how.

Way to go, Microsoft! {link to this story}

[Mon Jan 21 16:30:48 CET 2008]

David Pogue tries to explain in The new York Times why we seem to be witnessing a Macintosh surge. Pogue talks about the huge increase in attendance to this year's Macworld Conference & Expo, the increase in the amount of booths at the same event, more widespread use of the Safari browser as well as a serious increase in the amount of Macs sold last quarter (2.16 million). So, what could account for this increase in sales? I have to agree with Pogue that it has absolutely nothing to do with an iPod halo effect or the fact that users may be fed up with viruses and security problems in Windows. The key factor must be the Windows Vista fiasco. Just yesterday, I visited a good non-techie friend who bought an HP laptop the day before and was struggling to get Windows Vista to see the plain run of the mill wireless router over at her house. To make matters worse, every single time she tried to open any file or application, the OS presented her with a slew of pop-up windows with worrisome warnings about running applications insecurely and the possibility of getting viruses... all this on a system that hasn't even be connected to the Net yet! She plugged in a device she had been using before with Windows XP without any problem and the new OS refused to see it. Now she had to muck around with the drivers too, and don't even get her started on the fact that performance appears to be the same on this new and shiny laptop in spite of the fact that her old computer was nearly four years old. Shell a few thousand euros to end up getting the same performance as with your old laptop plus an OS that cannot be easily connected to the Net. Nice. So, what is causing people to switch to Apple then? Easy. By forcing users to switch to a different OS that also sports a new interface they have to get used to, Microsoft has provided millions of consumers out there with a wonderfult chance to try something new and get rid of all the Windows problems that have been driving them nuts for years. After all, once we internalize we need to learn a new interface, why not try something innovative, cool and easy to use that, on top of that, doesn't have nearly as many security problems as Windows? The only thing that could save Windows from losing a significant chunk of the market to Apple now (mind you, I'm not talking about 50% or anything like that, but rather something like perhaps 15-20% of the home market) would be Apple's own mistakes, lack of serious marketing outside the US and, above all, high prices. It turned out the open source advocates were right all along. The delays and problems with Windows Vista did provide a chance to diminish Microsoft's hold on the market after all. However, it's not Linux that has benefitted the most, but Apple. {link to this story}

[Thu Jan 10 15:03:49 CET 2008]

Interesting. A couple of days ago, I stumbled upon a document titled RPM Press Release announcing a new and improved version of RPM. Among other things, it talked about revamping the Automake build environment, porting the code base to several other major operating systems (I thought this had been done a long time ago, and I've known of an Irix port of RPM for quite sometime now), LZMA compression, etc. Well, little did I know that what I was reading was actually the press release published by the RPM5 folks, who led a fork of the official RPM project (yes, the title of the press release itself was quite misleading, perhaps even dishonest). In any case, I noticed about it thanks to an article by Sean Michael Kerner published in Internet News. For the time being, both products are fully compatible but there is no guarantee whatsoever that this will remain so in the near future. As far as I know, none of the major RPM-based distributions has switched to the new RPM5 system so far, whatever its advantages are. {link to this story}

[Wed Jan 9 15:27:03 CET 2008]

Interesting. Google Code includes a project called flyback that implements something similar to Apple's Time Machine for Linux users. It's basically a GUI written in Python on top of tools like rsync. {link to this story}

[Wed Jan 9 11:42:43 CET 2008]

How often do you come across an article so full of typical tired concepts that it almost hurts your eyes? I just read one such piece: The Hidden Costs of Linux Ownership, by Adrian Kingsley-Hughes. There are valid reasons to criticize Linux (some of them below), but this particular article is so full of typical outdated ideas that have been repeated for so many years now that one wonders if perhaps all the author did was put them together and pocket the money from the publisher. Here are some of the gems extracted from the piece:

The first cost is uncertainty. It's hard to measure in any definitive way but that doesn't mean that it should be ignored. When you take a copy of Windows XP, Vista or Mac OS X and you install it onto a system with the appropriate system requirements, chances are that unless you have a particularly bizarre configuration or a defective component, you can be pretty certain that the OS will install and things that you have installed (WiFi adaptors, network cards, graphics cards and so on) will work just fine.

Notice all the disclaimers included in the paragraph above. Things like "a system with the appropriate system requirements" or "unless you have a particularly bizarre configuration", for therein lies the key to the author's argument. As far as I know, this also applies to Linux, right? As long as the hardware is supported, it will work out of the box. The problem, of course, is that quite a few devices are not supported because the vendor never released the specs to the opensource community and proprietary drivers may or may not exist. In other words, the same problem one may run into with the commercial operating systems (drivers issues with Windows Vista anyone?). Another example: we recently bought an HP Color Laserjet 2600n printer that worked with my wife's MacBook Pro out o the box. Just plug it into the AirPort, find the device and print. However, it's quite spotty. Now it work, now it doesn't. Do we see any errors? No. Is the printer malfunctioning? No. Just plug the laptop to it directly via USB, and it works again. And these are all Apple products specificaly designed to work with each other. So much for Kingsley's "will work just fine". Want another example? My cousin's Windows XP's brainless firewall utility getting on the way of her Internet connection every few months. Does she fiddle with this or that? Nope. She is scared of it. Yet, every few months the thing stops working and prevents her form even connecting to the Internet, she calls me and I have to go check what's wrong. Again, so much for the "it just works". As for WiFi, I still have to see Windows XP working seamlessly with the AirPort in my house, or my wife's MacBook Pro consistently working with every single WiFi zone she finds out there in Spanish cafes, or... let's just leave it there. The reason why these things tend to work with Windows is just because people configure things for Windows. The truth is that today's technology (be it Windows, Linux, Mac or whatever else) is still quite finicky. The only thing that manages to improve things, but without completely solving the problem, is to stick with a single vendor for everything (the famous Apple experience, although then one has to deal with a different issue: the high price of the products.

Let's move onto the second argument from the article:

Another hidden cost is time. While it's true that installing Linux has become quicker and easier over the years, the process is still far from perfect. Some severe problem areas still exist (for example, WiFI adaptors, which is very hit and miss) and if you happen to run into the tar pits, you can expect to be stuck there for a long time. Also, there are more basic things. While Vista and Leopard are ready to play DVDs out of the box, Linux users have to mess about with codecs and agree to legally indemnify everyone for using legally dubious codecs. Sure, you can buy software players, some of which are rather good, but the advantage of a free OS starts to be eroded if you instantly have to put your hand in your pocket.

Kingsley is absolutely right when it comes to the DVD codecs, although not all the blame can be laid at the feet of Linux itself. To be fair, there are some complex legal issues going on here. If Windows and Mac OS were free, would they be able to give away the codecs too? He doesn't seem to be asking that question which, of course, leads him to completely miss the point when it comes to that last sentence. So, let me get this straight, paying hundreds of dollars for an operating system is OK, but paying a small amount of money for a multimedia app and getting an OS for free is not, right? Why does he think that the advantage of a free OS is "eroded if you instantly have to put your hand in your pocket"? In this sense, Kingsley appears to share the same point of view of the most rabid free software advocates, the ones who criticize Ubuntu for making it easier to download and install proprietary drivers (by the way, Ubuntu also makes it very easy to download and install said codecs without needing to do anything at the command line nor having to pay any amount of money whatsoever, but Kingsley seems to ignore that).

Anyways, things definitely go downhill from there:

Even support takes time as you have to post your request on a forum and wait for someone to come along and offer you a helping hand. The help is usually quite good and offered by knowledgeable people, usually giving their time away freely, but that still doesn't help you if you're working to the clock and you've promised someone that Linux is so much better than Windows. If you get a really obscure error message or particularly weird problem, you could be waiting for help for a long time.

Another element that consumes time is deciding which Linux distro to use. Many in the Linux community still refuse to accept that the number of Linux distros available acts as a barrier to adoption. Sure, choice is a good thing, and sure, it seems that Ubuntu is making good progress when it comes to becoming the mainstream "People's Linux" distro, and gOS has enjoyed success at the very-low-cost end of the PC market. But for the users who have heard the term Linux, narrowing down which distro is right for them might take time, especially if they happen to get caught up in a religious war between two distro factions.

Let's tackle this last issue first. I am one of those people who is convinced that choice is good, although I also acknowledge that it could introduce a certain measure of confusion among end users. However, what would Kingsley say of an open-minded customer out there who truly looks at the choices he has before purchasing an OS and realizes that there are, at the very least, four: Windows XP, Windows Vista, Mac OS and Linux? Gee, how complex is this! Am I expected to make a decision? This is so confusing! In other words, I think he is commiting a logical mistake: he is comparing the "confusing world" of Linux, where you can choose between different distros, to the supposedly straightforward world of Windows. He is wrong on two counts: first of all, I know people who are running Windows 98, Windows ME, Windows XP, Windows 2000 and Windows Vista. Unless someone takes the time to find out which is the latest and greatest Windows release, I'd say the level of confusion is quite large too (and I didn't even start talking about the options within each one of those: standard, home edition, business edition, cat lover edition...). Kingsley is not being totally honest. Besides, really, let's be clear, the only distros that still count as far as a regular end user is concerned are Fedora, SUSE and Ubuntu.

And what to say about the "support takes time" argument? Is it even worth spending any time on it? Did he ever try calling a technical support number? And did he ever get the problem solved within 10 minutes? What if he gets "an obscure error message or particularly weird problem" in Windows or Mac? How much time does he have to spend to get it fixed? Even more to the point: how much money does he have to pay to get some decent support? {link to this story}

[Fri Jan 4 15:18:34 CET 2008]

I have come across an old article about Jaron Lanier published by The Guardian that makes an interesting read. For all those who've never heard of him before, Lanier (check his personal website or the Wikipedia entry) is a pioneer of the concept of virtual reality. Incidentally, there are a couple of paragraphs from the article on this particular topic that are worth quoting here;

"If you run into somebody at a cocktail party now, they'' think of VR [Virtual Reality] as an escapist medium, and I know why it happened. In the 1980s, Tim [Leary] made his living on the speaking circuit and he got this idea that virtual reality would be a striking thing to talk about. At first glance, it does seem to have something to do with the psychedelic experience. But it's almost the logical opposite. You're not in retreat, you're not passive, you;re not having an experience wash over you. You have to be intentional. You get tired. It's a waking state activity. It's not like taking drugs: it's like going on a hile. Or "—he pauses to construct a characteristically elaborate metaphor— "it's like going on a hike and being the sculptor of the mountain at the same time".

That's precisely the mindset that brought us the Wii. I fully agree with Lanier that way too many people out there have the wrong idea about virtual reality and computing in general, viewing them as passive acitivities. They may be so... or not, depending on what you choose to do with them. They are just tools. {link to this story}

[Fri Jan 4 14:48:32 CET 2008]

Computer World has published a good piece on the top 10 personal tech trends for the year 2008: flash-based superportables (by the way, the Asus Eee PC looks like a great laptop to introduce kids into the world of computing), free wifi access hardwired into products, home robots, hyperconnectivity, multi-touch interfaces, the unstoppable spread of GPS devices, reading on-screen, the "social everything" (interesting aside about Google's OPenSocial API), etc. Interesting read. {link to this story}

[Fri Jan 4 14:30:38 CET 2008]

Today I came across a reference to the IEs4Linux project, which I had never heard of. It makes it easier to install Internet Explorer on a system running Linux. The installation instructions will help you get it up and running on any of the major distributions: Ubuntu, Fedora, Gentoo, Debian, SUSE, Mandriva... It's obviously geared towards web developers who need to test their websites on Microsoft's browser, but I imagine there may be some people out there who'd rather run MSIE than Firefox. Weirdos. {link to this story}

[Wed Jan 2 12:41:07 CET 2008]

Rob Rohan writes about his failed attempt to switch from Mac to Ubuntu Linux and, interestingly enough, reaches a conclusion that goes beyond the mere criticism of Linux on the desktop that we see so often:

The one thing I learned from this, is I am currently locked into Apple stuff. I didn't quite realize it until now though —it never really hit home. For example, if I moved back to Linux I would have no way of syncing my iPhone. I also have an Airport Extreme, and near as I can tell, I can't edit any of the Airport settings from Linux. Let alone my iTunes purchases. Wow, I am toast.

I always imagined I could just install Ubuntu on my Apple hardware if it ever got down to it, but it seems that it is not that easy. By the way, hats off to the open firmware guys and yaboot —I was impressed that Ubuntu actually could install on and support most Mac hardware.

f you are running Intel hardware, you are not likely in this same boat. You can probably use boot camp to easily dual boot or use vmware.

My lesson mearned here is —single vendor lock in is limiting on your personal maneuverability, no matter how cool the products seem to be. If the company you are hooked in with starts putting out stuff you don't like, the economy takes a dump and you can't afford their stuff, or you disagree with something, your options are limited.

Rohan decided to give Ubuntu a try after he upgraded to Leopard and ran into way too many issues (applications crashing all the time, admin privileges removed, problems to sync his iPhone, etc.). To be fair, my wife upgraded to the same version of the OS and everything worked flawlessly. She likes it, and so do I whenever I have the chance to work on her laptop. Apple does make cool products and, right now, it's light-years ahead of Microsoft. There is little doubt in my mind the latter still dominates the industry due to its sheer size, not to the quality of its products or its innovation. However, cool as they are, Apple products, like any other, also break, fail to do what one has in mind or needs and, of course, they also run out their commercial life and are thrown to the dustbin. When that happens, the end user is left out in the cold. This happened to Rohan with his Power Mac G4, for example. Soon he won't be able to run any new Apple OS on it, and the old ones won't be supported anymore. A company made a decision to get rid of a given computer model that still works fine for him and he will have no other choice but to shell out the 2,000 bucks to buy a new and shiny Apple, whether he truly needs it or not. Now, that's not so different from being locked in by Microsoft, right?

Knowing that my wife uses Apple products and that I do like the way they are designed and run, a good friend of mine recently asked me if I'd recommend purchasing a Mac for home use over a PC running Linux. Well, it depends. If you are a gamer, you still need a Windows system. Will Boot Camp help? Perhaps. I didn't test it yet. In any case, to run Boot Camp you need an Intel based Mac. If you need a multimedia system that can also help you do some graphics design, compose your own music, view videos and manage your large collection of pictures, buy an Apple. If you just browse around, use email and write documents, buy an Apple. However, if you value your freedom above all, don't like the idea of using pirated software because you cannot afford their high prices, are technically savvy and/or need to do remote system administration and programming, then use Linux. It still is more flexible for that type of work and it comes with lots of free apps. In addition, there may also be some philosophical reasons to bet on Linux versus Apple that we all know quite well. One way or another, what I don't understand is how anyone could be still running Windows for a reason other than pure inertia or "it's what everybody else does". If one's priority is usability, Apple outperforms both Windows and Linux without any doubt; if one needs flexibility and customizability, Linux is the winner of them all, although Apple comes a close second; for multimedia and graphics capabilities, Apple is once again the winner... as I say, I'd only use Windows if I were a big gamer, which I'm not. Other than that, it's perfectly possible —even advisable— to do without Windows. {link to this story}

[Wed Jan 2 08:50:40 CET 2008]

Well, we all knew it had to happen sooner or later. AOL has decided to end the development of the Netscape browser after years and years of desperate survival in the shadows. The once mighty browser, the one that started it all —at least in most people's eyes, since NCSA's Mosaic is the one that truly deserves that title when it comes to graphical browsers— will not be developed anymore starting in February 2008. Some way to start the new year! I only can say that I have plenty of good memories from the couple of years I spent supporting the product from DecisionOne. We were all pumped up, excited to be supporting a cool product that was changing the world. I can say without any doubt that Netscape is what got me interested again in the world of technology. Without it, I wouldn't be here now. I find it interesting that even the Spanish news covered the end of Netscape, in spite of the fact that most people over here don't even know what it is. By the time it made it here, the Web was a synonim of Internet Explorer, the browser that came with the Windows OS by default. When I visited Spain back in 1997 or so, it was clear in my mind that Netscape was truly doomed. As the Web was expanding through countries other than the US, users only knew Internet Explorer as their gate to the new world. Nobody had heard of Netscape or any other browser. The die was cast. {link to this story}