Author Archive

Comentario en enriquedans.com

Transcribo el texto de un comentario que he mandado al blog de Enrique Dans, y que por algún motivo no ha salido correctamente (lo voy a enlazar desde allí, para ver si sale así):

¡Cuánta ignorancia junta, Dios mío!

Dice #15:

“Para alabar a Beryl tiene uno que probarlo, no poner un video del youtube… Beryl esta verde, verde, verdisimo… “

Mentira. Yo llevo tiempo usando Beryl, y ha madurado mucho en muy poco tiempo. En mis ordenadores “de trabajo” no lo uso, porque soy más bien de los que tienen escritorios espartanos: Xfce con un solo panel abajo y un fondo de escritorio sin iconos. Y sin efectos 3D ni transparencias.

Ahora bien, cuando puse Beryl por primera vez, me funcionó de maravilla, y podía hacer un montón de cosas que no sé si Aero puede hacer (igual sí). Por ejemplo: Alt+Rueda del ratón sobre una ventana y cambio su transparencia de 0% hasta casi 100% (y puedo ver lo de debajo). Puedo ver dos o tres vídeos diferentes a la vez, con diferente nivel de transparencia, y ver los de abajo a través de los de arriba. Y eso mientras pongo un efecto de lluvia sobre todo el escritorio. Y eso mientras roto el cubo, o pongo las ventanas con los vídeos en una arista del cubo…

Y todo esto antes de que Vista saliera al mercado, y con ordenadores en los que Vista no funcionaría, porque “su alta tecnología requiere mejor hardware”.

¿Para qué sirve esto? Para nada. Igual que Aero. Simplemente mola, y si lo quiero usar, puedo hacerlo. Yo, la verdad, no lo uso, pero para gustos los colores.

Pero el escritorio Linux no se acaba en Beryl. Hace años que en funcionalidad el escritorio Linux (GNOME, KDE, Xfce… incluso Fluxbox y similares) ha sobrepasado ampliamente a cualquier Windows, incluido el Vista. Desde escritorios múltiples, hasta shortcuts de teclado ultrapersonalizables, colocación de ventanas automática más inteligente (al abrirse una aplicación), lista de tareas más eficiente, paneles configurables, “gadgets” (como los han rebautizado los sinvergüenzas de Redmond) como relojes y monitores gráficos de uso de CPU, red o I/O de disco etc. sobre el escritorio.

Por no mencionar la brutal capacidad de personalización de la interfaz, cambiando la decoración de las ventanas, el estilo de los botones, menús, listas, etc, la fuente de letra para los títulos de las ventanas, los menús, los iconos…

Leo en #37:

“Luego tengo que hacerles una aplicación cliente, sin que tenga que tirar más lineas de código que un tonto, y los dos únicos lenguages que tengo disponibles (no hay Visual Basic) son C y Java.”

Pero bueno, chaval, ¿estás diciendo que no usas Linux porque no hay lenguajes de programación? Si quieres aplicaciones para cálculo masivo tienes Fortran y C. Si quieres scripts rápidos, eficientes, y fáciles de coj*nes de hacer, tienes shell, Perl, Python, Ruby y otros. Si quieres aplicaciones gráficas fáciles hace tiempo que tienes Tcl/Tk, y más moderno GTK+ y Qt. Tanto Tk como GTK (Qt no sé) tienen una integración con Perl , Python y C que asombra por su simplicidad.

Me gustaría ver que programa “simple” de VB u otra basura similar es capaz de hacer lo que dos líneas de Perl o shell, con sed y awk. Para que te hagas una idea, Google usa Perl para pattern matching cuando te da los resultados de una búsqueda. Sobre máquinas Linux, claro.

Date una vuelta por la Wikipedia, y su lista de lenguajes de programación por categorías, y verás la de lenguajes diferentes que hay, y haz la cuenta de cuantos se pueden usar en Windows y cuantos en Linux.

Luego dice #44:

“Eso de que quien usa Windows es porque quiere no es verdad. Yo llevo años intentando emplear Linux y no hay manera. Hace siete años lo probé por primera vez con una distro de Mandrake y me volví tonto.”

Quizá no sea justo culpar a Mandrake de esto último…

Es laudable tu intención de usar Linux, y lamentable que no lo hayas conseguido, pero creo que tu negativa experiencia no es necesariamente generalizable.

Yo llevo 9 años usando Linux. Empecé con Slackware, donde uno se hacía todo “a mano”. ¡Qué tiempos! Era complicado a veces, pero aprendí muchísimo. Luego, cuando probé Mandrake, me gusto mucho, porque era tan fácil que hasta daba un poco de vergüenza.

Con el tiempo, volví a distros más “técnicas”, y ahora uso Debian (que es como el “Ubuntu para frikis”), porque me permite más flexibilidad que las distros “para tontos” (con todos los respetos), y me es mucho más fácil controlar lo que hace el ordenador, que con distros que se creen más listas que yo, y me “facilitan” el hacer las cosas como creen que quiero hacerlas, y no como quiero hacerlas.

“Desde entonces lo he vuelto a intentar varias veces y siempre me he topado con un muro de piedra: la conexión a internet.

JAMÁS he logrado conseguir conectarme a internet con una distribución de Linux. Ni cuando usaba un módem RTB, ni usando un módem ADSL, ni ahora con un router wifi.

Pues debes de ser el único, macho. Yo tuve problemas con el v.90/92, cuando intente conectarme con un winmódem interno. Pensé que Linux era una castaña, hasta que me compré un módem externo, y vi que era IGUAL de fácil de configurar que en Windows (y más fiable).

Cuando me pasé al ADSL (en realidad tengo cable, con Euskaltel), no tuve NINGÚN problema con Linux. Lo configuré en un tris. Y cuando me puse WiFi, me compré yo mismo el router (con lo cual me ahorré unos eurillos, respecto a pedirlo a Euskaltel), y me lo instalé sin problema en el de sobremesa (con cable). El portátil que conecto por WiFi no me ha dado ningún problema para conectar en modo abierto, y tampoco con encriptación WEP. Cierto es que para WPA tuve que hacer alguna cosilla, y que en Windows es más sencillo, pero solo marginalmente más sencillo.

Como comentario final, añadir que Vista no hace más que reinventar la rueda, reimplementando mil cosas que ya existían en Mac y en Linux (y generalmente, mucho mejor hechas), y cambiándoles el nombre, para que parezca que las han inventado ellos (como muestra un botón: los infames “gadgets”, que son el último S.O. del mercado, libre o no, que los implementa, y lo venden como que fueran los inventores).

Comments (2)

Inkscape tip: make arrow head’s color match that of its body

I have encountered the problem more than once, and it is a bit annoying to say the least. Basically, when you build a path/arrow in Inkscape, it starts as a black curve by default. You can edit it to put a marker in either or both ends (Click on the curve, then Object->Fill and Stroke->Stroke Style), to make an arrow, for example.

Now, the problem is that if you change the color of the body of the arrow, the head will remain black, as documented, for example, in A Guide to Inkscape, by Tavmjong Bah. Not nice, uh? The solution is given in the same site, and consists on using a plugin. To do so, select: Effects->Modify Path->Color Markers to Match Stroke.

If you are a Debian user, you might encounter a problem: a window pops up saying The inkex.py module requires PyXML. This has been reported as a bug, and also happens for Ubuntu. The solution is to install the python-xml package, which is not always installed by default when you install Inkscape, it is just “suggested”. This means that when you install Inkscape (aptitude install inkscape), aptitude will tell you something like “The package python-xml is recommended, but it is not going to be installed”, and will go on happily. If (like me) you ignore the suggestion, you will not have the python-xml package installed, and some extensions, like the above, will not work (however this allows the users that do not want to use the plugins to have a lighter instalation, if they so wish).

Comments (18)

Prisoners, queens and why we should bother about them

There are two concepts that I find very interesting, and that apply to many situations, from everyday life to international politics. One concept applies to prisoners, and the other to Royalty, but their long arms reach much farther.

The “royal” concept is that of the Red Queen’s race, taken from Lewis Carrol’s Through the Looking-Glass (aka “Alice in Wonderland II”). This race is one in which runners must run the fastest they can to stay in place. To move, they’d have to run twice that fast.

The other concept is that of the prisoner’s dilemma. The dilemma is a game with the following rules (taken from Wikipedia):

Two suspects, A and B, are arrested by the police. The police have insufficient evidence for a conviction, and, having separated both prisoners, visit each of them to offer the same deal: if one testifies for the prosecution against the other and the other remains silent, the betrayer goes free and the silent accomplice receives the full 10-year sentence. If both stay silent, both prisoners are sentenced to only six months in jail for a minor charge. If each betrays the other, each receives a five-year sentence. Each prisoner must make the choice of whether to betray the other or to remain silent. However, neither prisoner knows for sure what choice the other prisoner will make. So this dilemma poses the question: How should the prisoners act?

The “ideal” solution would seem both to stay silent, but if you look closely, whatever the other player chooses, any single player is better off by betraying. So, any rational player should choose to betray, even though this leads to both betraying, and gives an overall lower payoff (higher punishment for both).

Now, for the third part of the title of this post… why should we care? Well, it seems to me that we can find all around us cases of Red Queen races caused by sub-optimal solutions to prisoner’s dilemmas. For example, it is quite apparent the rise in SUVs and all-terrain vehicle sales. People here in Europe seem to start following the silly North-American custom of buying the biggest vehicle available, regardless of usability and needs fulfilled. One of the pseudo-reasons given by sellers is that SUVs are safer. Why would that be? Well, because if a small car and a SUV crash head-on, the passengers of the latter are much more likely to be less hurt than the ones in the former. This sounds rational… but is utter crap. I don’t claim that people buy these vehicles for that reason, but it helps.

Now, let’s analize the scenario: it is true that in a SUV/small car crash, the SUV is better off. However, SUV/SUV crashes are worse for all passengers than small car/small car hits are. From that information, it is apparent that we are facing a prissoner’s dilemma (not counting the fact that SUV/wall hits are also worse). Buying a SUV would be betraying, and buying a small car cooperating. The buyer of a SUV hopes that all other players/buyers get small cars, so that her option gives her an edge over the others. However, if we all think the same, we’ll all buy SUVs, and then we will reach a betray/betray equilibrium, when a coop/coop equilibrium would be better for all. We’d be running a Red Queen race, only to end up in the same place: all with SUVs, instead of all with small cars… but all with equivalent vehicles (and actually worse, overall).

Another similar situation would be that of the arms race. We all know the story: two or more countries/factions increase their weaponry, not to be overwhelmed by the other country/faction, in a potential war. Now, no matter what country A does, country B will be better off stocking more weapons: if A stays unarmed, B can beat it. If A arms itself, B has to arm itself not to be beat. However, Both countries being armed (betray/betray) is immensely worse than both countries being unarmed (coop/coop). In both cases the war is deterred by the offensive/defensive equilibrium, but in the former the risk for a catastrophe is much higher.

We are fooled by governments and army leaders, assuring us that other countries will play the “betray” card (and arm themselves), so we should play it too. However, think of the fact that in their countries, the other citizens are told exactly the same about us by their government. An no-one seems to explain that the betray/betray solution is sub-optimal, and that coop/coop solutions could exist.

I have no solution for these issues… but, dear reader, maybe you could find it if you thought about it. Please, do.

Comments (2)

Happy Wikibirthday to myself

As of today, I have reached my first year of (English) Wikipedia contributions under the identity isilanes. I have accumulated over 2100 edits in that time, and created over ten new pages, not counting dozens of pictures of chemical structures I have made for chemistry articles.

Other contributions could be the logo of my University (UPV/EHU), that appears at its wikipedia page, many vandalism warnings to misbehaving users and disambiguation of links.

Comments (1)

Some changes in the blog

The constant reader, if such there be, might notice that I have made some small changes in the design of the blog. I have made the right sidebar wider, and changed the radio.blog I had there for one of the Jamendo player widgets. It features the music of groups I like from Jamendo, and will be updated automatically when I add “preferred” groups to my stats at Jamendo.

Happy listening, and I hope you don’t find the modified layout worse than the previous one.

Comments

I cross the 2k edit mark in Wikipedia

Following my recent trend of using the blog to talk about myself, instead of about the FLOSS that gives name to the blog, and after the Wikipedia fever I talked about some days ago, I write again to say that I have surpassed the 2000 edit count in Wikipedia, as you can see following this link.

Still, as crazy and monomaniac it sounds, guess what category of Wikipedia I belong to, regarding edit count? “Prolific editors”, maybe? Or “Wikipedians with over 1000 edits”? Or “Top X Wikipedia editors”? No! I belong to “Wikipedians with fewer than 5000 edits” :^(

Comments

App of the week: Eye of GNOME

I recently discovered this little application, and I must confess it nicely fits a niche. The Eye of GNOME (eog), is a kind of clone of the Windows default picture viewer, and is a good complement for other Linux tools like ImageMagick.

I use the display tool of the ImageMagick package for highly repetitive and/or precise transformation image watching (as in putting one image above another, then watching the result, then making the composition again if it was not OK, or resizing a set of images to a given exact percent of their original size).

On the other hand, eog is nice for watching a lot of images in a row, and having them automatically resized to fit in the watch window. eog also permits smooth scrolling with the mouse wheel, very fast image rotation, and single-click window fitting of the image.

Give it a try!

Comments (2)

Wikipedia fever

Yesterday was the last day of May, and an incredibly fruitful month ended, regarding my Wikipedia edits. As you can see in the previous link, I was 75 edits short of making 1000 edits that month! Approximately one half of all the edits I have done in Wikipedia in one year (in 18 days it will be the birthday of my first edit as Isilanes), where done in May 2007!

This huge (for me) amount of edits was possible due to the kind of activity I have had in Wikipedia as of lately. I realized there are a lot of chemical structures in Wikipedia that are of low quality, and I started to improve them. The main flaw of low q images is that they are done in raster format (PNG, GIF, JPEG). This implies that they lose quality upon magnification, and that the larger they are, the more space they occupy. In contrast, vector graphics (e.g. SVG) offer a perfect quality regardless of magnification, with a constant file size, no matter what output size we ask of them.

Actually, there is a page in Wikipedia specifically devoted to listing the (chemical) images that, due to being easily translated to SVG (most chemical skeletal formulae fall in this category), and showing a low quality, are suitable for being substituted by SVG counterparts.

Once I found this page, I started using the superb free software programs ChemTool and Inkscape to draw SVG counterparts for many structures. For each structure, a file has to be uploaded, the article including the raster image has to be modified to include the SVG instead, then the raster image has to be tagged as already superseded by a SVG, and depending on the case, it can be tagged with a proposal for deletion. In that case, it has to be included in a page listing the images and media for deletion, and the original uploader of the image should be notified, as a polite measure. This implies a bunch of edits per superseded image.

A great part of my May edits also correspond to the fact that I modified a previous SVG image that was to replace a PNG of some Free Software logos. Apparently the “old” SVG had some errors, which I corrected:




PNG version of the SVG image

This logo picture was used by a tag that appears in all articles related to free software, and I starting changing the appearance of the old pic in every tag with the new pic. The result: a whole lot of edits.

Comments

My uptime hits 60d

Following the irrelevant custom begun with a post two an a half months ago, I’ll share my uptime data with the world again.

Short after my computer hitting 50d uptime (at the time of my previous post), I had to reboot it (I must admit it’s one of the 3 or 4 times in ten years that I had to reboot for technical reasons), so now it is “building up” uptime again. As of the writing of this post:

% uptime
10:55:22 up 61 days, 18:56,  1 user,  load average: 14.22, 14.37, 14.27

I fear I will have to shut it down sometime soon, because I will be moving to a new lab (the DIPC). I don’t know if someone will inherit my old computer and/or if it will be reformatted. Anyway, I’ll post every time a 20d milestone is hit (yes, just because I feel like it. Last time I checked, this was my blog. Got a problem? Sue me).

Now back to work.

Comments (1)

LaTeX: the textpos package

I have recently discovered a marvelous LaTeX package, namely textpos (PDF manual). My problem was that I wanted to place pieces of text arbitrarily in a page. I had tried tabular environments, a liberal use of \vspace and \hspace, the minipage environment, and even the picture environment. None would do.

However textpos is just what I wanted. The package is loaded with the usual:

\usepackage[options]{textpos}

The options I have used are:

  • absolute – the placing is made with respect to the whole page (stuff me if I understand this)
  • showboxes – draws boxes around the text, so you exactly see what you do (for debugging, of course)

Some variables have to (can) be set, namely the width and height units, for example:

\setlength{\TPHorizModule}{\paperwidth}\setlength{\TPVertModule}{\paperheight}

After that, all the placing specifications can be referred to \TPHorizModule and TPVertModule. A typical use of textpos would be:

\begin{textblock}{width}[xt,yt](X,Y)  the-text-goes-here  \end{textblock}

where width is the desired width of the text box (the height will be enough to place all the text specified), X and Y are the (x,y) placement of the text box, and xt and yt are the point inside the text box which will be placed at (X,Y). All the units refer to \TPHorizModule and TPVertModule. For example:

\begin{textblock}{0.6}[0.5,0.5](0.3,0.4)  hello world\end{textblock}

will print the text “hello world” in a box of width 60% of \TPHorizModule (in my example, this is 60% of the total page width). The center (0.5,0.5) of that box will be placed at a point 30% to the right of the left margin, and 40% below the top margin (in TPxxxModule units).

Comments (1)

« Previous Page« Previous entries « Previous Page · Next Page » Next entries »Next Page »