Ayuda al desarrollo del DDR de CTIC

Queridos todos: La Unidad de Movilidad e Independencia de Dispositivo del Área de I+D+i de Fundación CTIC Centro Tecnológico está preparando un Repositorio de Descripción de Dispositivos distribuído y colaborativo y os pide ayuda a todos. Por cierto, nuestro sitio web está en medio de algunos cambios serios así que disculpas si hay algún problema con alguna dirección de las anteriores o si hay alguna sección en inglés que no funcione correctamente.

¿Qué es un Repositorio de Descripción de Dispositivos? (DDR)

Básicamente, es una base de datos de dispositivos que almacena información sobre las capacidades hardware y software de cualquier dispositivo. Puedes interrogar al DDR para intentar averiguar qué fabricante hizo el dispositivo, qué versión de sistema operativo lleva, qué navegador web tiene instalado, qué sensores tiene disponibles (brújula, acelerómetros, cámara, micrófono, ...), modelo de CPU, memoria principal y secundaria instaladas, almacenamiento disponible, su tamaño físico de pantalla, su resolución en pixels, la versión de Bluetooth y perfiles disponibles, codecs de audio y vídeo soportados, y muchas otras cosas. Principalmente, casi cualquier cosa que puedas saber del dispositivo de antemano, puesto que los DDRs mantienen información estática que no varía a lo largo del tiempo. Hay una muy buena entrada en la Wikipedia, si quieres indagar más sobre el tema.

¿Cómo puedo ayudar?

Normalmente, los DDRs reciben evidencias sobre la identidad del dispositivo y una consulta acerca de una o más características del dispositivo. A partir de esto, responden con valores para las capacidades del dispositivo requeridas en la consulta. Éste es sólo uno de los múltiples casos de uso para un DDR, pero considero que puede ser suficiente para que tengáis una idea acerca de las posibilidades de esta tecnología. Por ejemplo, cuando accedo un sitio web, el navegador de mi dispositivo (mi PC de sobremesa, mi portátil, mi teléfono móvil, mi televisión, mi set-top-box o mi consola de juegos) envía una Petición HTTP, analiza todas las cabeceras que ésta incluye y añade el dispositivo al DDR que estamos preparando. NECESITAMOS QUE ACCEDÁIS A http://idi.fundacionctic.org/mobHeaders O AL MÁS CORTO http://bit.ly/headrs CON TANTOS DISPOSITIVOS Y NAVEGADORES WEB COMO OS SEA POSIBLE. Queremos tener accesos de dispositivos reales pues no tomaremos información de tecnologías DDR existentes como WURFL O Device Atlas. Básicamente, porque no es ético aprovecharse del trabajo de otros y, en ocasiones, tampoco es legal.

¿Y qué saco de esto?

De momento, una respuesta agradeciéndote la contribución más una frase indicando si ya se ha añadido un navegador o dispositivo idénticos o si uno igual ya había accedido al sitio. Después de tener suficientes accesos y nuestros algoritmos más refinados, enriqueceremos el sitio web ofreciéndote información sobre las capacidades de tu dispositivo. A más largo plazo, ofreceremos una tecnología DDR gratuita con todos sus despliegues compartiendo información sobre identificación de dispositivos y sus características. Nuestra intención es acabar con la actual situación de jardines vallados en identificación de dispositivos. Creemos que ya debería haber una base de datos de dispositivos global y universal lista para todo el mundo, de forma que la comunidad de desarrolladores pueda centrar sus esfuerzos en la creación de software adaptativo de alta calidad basado en una fuente de identificación de dispositivos de confianza. Inicialmente, este DDR cubrirá adaptación web pero, si tiene éxito, ¿quién sabe qué más puede llegar? Tenemos algunos ases en la manga que compartiremos en cuanto estén maduros.

¿Cuándo estará listo?

No tenemos prisa. Tenéis excelentes soluciones comerciales como Device Atlas o WURFL, así que estás cubierto por buenas bases de datos de dispositivos en el dominio de aplicaciones y contenidos web. Antiguas versiones del API y base de datos de WURFL son gratuitas y de uso libre bajo la licencia correspondiente y pueden ser útiles para que desarrolles tu propia base de datos de dispositivos. El OpenDDR, con una aproximación similar a la nuestra, ha sido recientemente liberado. Sin embargo, podemos decir que habrá una primera versión en 2012 (si la profecía maya no nos para) y es muy posible que tengamos una versión anticipada en la primera mitad de este año.

Por favor, no nos hagas trampa

Sí. Sabemos que algunos de vosotros os sentiréis tentados de lanzar algunas peticiones HTTP por programa, usar software que permite trucar el agente de usuario del navegador, inundar nuestra base de datos con peticiones falsas, comprobar la seguridad del sitio y demás. Sabed que retrasaréis nuestros resultados y será una pena pero no podemos hacer nada contra eso. Sabemos que puede ocurrir y cargaremos con esa posibilidad. Estarás molestando a desarrolladores con el objetivo de proporcionar nueva tecnología gratis a la comunidad de desarrolladores. Ésta es simplemente una herramienta interna para que nosotros cataloguemos todos los dispositivos que pasan por nuestras manos y sólo pedimos ayuda en el proceso.

Vale, contribuiré. ¿Qué hago? ¿Cuándo paro?

Simplemente, accede a una de las direcciones web disponibles, (http://idi.fundacionctic.org/mobHeaders o la más corta y mejor para móviles y otros dispositivos con los que es complicado escribir http://bit.ly/headrs) con cada navegador web al que tengas acceso. Si actualizas la versión de sistema operativo de tu dispositivo y/o del navegador web, por favor, recuérdanos y accede de nuevo.

Sigue sintonizándonos

Sí, llegarán pronto más noticias aquí o en el sitio web de CTIC. Estamos ansiosos por traer nuevas noticias pronto. Mientras tanto, iremos enriqueciendo nuestra base de datos con todos los dispositivos que lleguen a nuestras manos. ¿Nos ayudarás? ¡Gracias!

Help developing CTIC's DDR

Dear all: The Unit of Device Independence and Mobility of the R&D Department at Fundación CTIC Centro Tecnológico is preparing a distributed collaborative Device Description Repository and is asking for help to you all. By the way, our web site is in the middle of some serious changes so sorry if no English version is available or there is some problems with any of the web address above.

What is a Device Description Repository (DDR)?

It is basically a device database which keeps information about hardware and software device features. You can query a DDR in order to guess what vendor made the device, what the operating system version is, which web browser is installed, what sensors are available (compass, accelerometers, camera, microphone, ...), CPU model, primary and secondary memory installed, available storage, its physical screen size, its resolution in pixels, Bluetooth version and available profiles, audio and video codecs supported, and many other things. Mainly, mostly everything that you can know beforehand about the device, as DDRs keep static information which does not change over time. There is a good wikipedia entry if you want to dig more on this.

How can I help?

Typically, DDRs receive evidences about the identity of the device and a query for one or more device features. Consequently, they respond with values for the device features requested. This is only one of the multiple use cases for a DDR, but I consider it is sufficient for you to guess their purpose. For example, when I access a web site, the web browser of my device (my desktop PC, my laptop, my mobile phone, my TV set, my Set-Top Box or my game console) send a message called HTTP Request which includes some information which helps identifying which device/browser performed such request. So we have prepared a very basic web site that you can access with any of your device and web browsers. In this way, we shall read the HTTP Request, analyze all the headers included and add the device to the Device Description Repository that we are preparing. WE NEED THAT ALL OF YOU ACCESS http://idi.fundacionctic.org/mobHeaders OR THE SHORTER http://bit.ly/headrs address WITH AS MANY DIFFERENT DEVICES AND WEB BROWSERS AS POSSIBLE. We want to have real hits from real browsers because we will not sniff information from existing DDR technologies such as WURFL or Device Atlas.

And what do I get from it

So far, a response thanking you for contributing plus a sentence indicating if an identical web browser and/or device had already hit the site or if you are the first one which contributed those device identification evidences. After we have a sufficient number of devices and we have refined our algorithms, we shall enrich the site providing you with some information about your device. In the long term, we will release a free DDR technology with all the deployments sharing device identification and device features. Our intention is to finish the current state of walled gardens in device identification. A global universal device database should be ready for everyone so the developer community can focus their efforts creating high-quality adaptation software based on a trustworthy device identification source. Initially, this DDR will cover web adaptation but, if successful, who knows what might come? We have some aces in the hole which we shall share when they are mature.

When will it be ready

We are not in a hurry. You have excellent commercial solutions such as Device Atlas or WURFL so you are covered for good device databases in the web domain. Old versions of the WURFL API and database are free and they might be very helpful for you to develop your own device database. The OpenDDR, with an approach more similar to ours, has recently been released. However, a first version will be ready in 2012 (if the Mayan prophecy does not stop us) and it is very likely that we have an early access version in the first half of this year.

Please do not play tricks on it

Yes, some of you will feel tempted to make some HTTP Requests programatically, use user agent string faking software and browser extensions, flood our database, check security of the site and so on. Know that you will delay our results and it will be a pity but we won't do anything against that. We know it can happen and we shall bear with that possibility. You will be disturbing developers with the aim to provide new technology for free to the developer community. This is just an internal tool for us to catalogue all the devices which come to our hands and we are asking for help in the process.

OK, I will contribute.- What to do? When to stop?

Just access one of the available web addresses (http://idi.fundacionctic.org/mobHeaders or the shorter good for mobile phones and other hard-to-type-with devices http://bit.ly/headrs) with each web browser that you have access to. If you update your device OS version and/or browser, please remember us and access again.

Stay tuned

Yes, more news will come soon here and/or at CTIC's web site. We're looking forward to bringing good news! In the meantime we shall enrich our device database with all the devices that reach our hands. Will you help us? Thank you!

Box.net client for Mac OSX

I own a paid account in Dropbox and I do love it. The extra Rat-Pack allows me to have infinite versions of the file in my Dropbox. I have already accidentally deleted important files more than once and the Dropbox web client allowed me to undelete it. The 50 GBs of the account are great, although if you intend to keep back-ups of old email inboxes and some other old stuff, you can fill it. It has happened to me. Then came box.net with the offer of 50 GBs to iOS users in order to counteract the iCloud offer of 5 GB for free. So I signed up and I checked that there is no native client as there is for Dropbox. I saw people complaining about it and it seemed weird to me that there were Mac OSX users complaining. Why? Mac OSX can use Finder in order to access their box.net files via WebDAV. You just have to bring Finder to the foreground, press cmd+K to connect to server and then write the URI to connect to Box's WebDAV: https://www.box.net/dav/. Then type in your login and password and there you have it. So I have the Dropbox folder which keeps space in my hard disk, some Dropbox folders which are in the cloud (thanks to Dropbox's selective sync feature) and all the files in Box.net remotely accessible thanks to WebDAV and its Mac OSX's finder integration.

Converting NRG disk images to the ISO format in MacOSX

First step: install MacPorts (Direct access to the latest .dmg file for OSX Lion at the very moment of writing this post is here). MacPorts is a set of software tools which allow distinct open source software from other POSIX-like systems such as Linux. These open source software pieces include command-line tools and X11 tools. It includes a software packaging system very easy to use. So easy, that, we only need to install nrg2iso by executing the command

sudo port install nrg2iso

in Terminal. Then, just execute

nrg2iso inputfile.nrg outputfile.iso

in Terminal's command line. Two things may happen afterwards: console output may show the advance in the task of converting the original Nero .nrg file to the more standard .iso disk image format; alternatively, the .nrg file may already be a .iso file simply renamed to .iso and console output will warn us about that fact. In this case, we only have to change the .nrg extension to .iso.

Survey illustrates iPad's effect on PC market

Tuaw has just published a post about how PC users spend less time after purchasing an iPad. It references a survey by AdMob and they say that "Among the 1,430 tablet own ers who par tic i pat ed in the sur vey, 77 per cent admit ted to spend ing less time with their per son al com put ers after buy ing a tablet, 43 per cent said they now use their tablets more than their per son al com put ers, and 28 per cent claimed to use a tablet as their pri ma ry com put er. Most tablet users, 68 per cent, spent at least one hour each day with the device, pri mar i ly for play ing games (84 per cent), search ing for infor ma tion (78 per cent) or email cor re spon dence (74 per cent)."

It must be true ;-)

Written from an iPad ;-)

Trying to post from Blogger app in Android

So that is it. Just my first try blogging from the Android Blogger app.

Microblogging killed the blogging star

Ain't that true? ;-)

New social network item: gdgt gadget list in the sidebar

You can see a new element in this blog's sidebar: my gadget list in gdgt. This is an awesome social web site specialized in gadgets. Get there and try it :-)

Solid State Disks: An important improvement for a computer system

I have just replaced the original Seagate Momentus 7200.2 320 GB of a Dell Studio XPS 1340 laptop with a Kingston SNV 225-S2/128GB from the SSDNow V+ Series.

The new SSD disk allowed me to install Windows 7 64-bit in 15 minutes. This time includes the accumulated time spent in the final installation screen of the Windows 7 Install CD (from the "Copying Windows files" step to the "Installing updates" step, which was 10 minutes and 05 seconds), in the "Completing install" (3 minutes) and the final "Preparing computer for first use" (which took less than one minute). The total accumulated time also includes two computer reboots needed in the process: unluckily, the BIOS is quite slow in the boot process so each reboot takes about 30 seconds, withouth the disk being the one to blame.

Some other interesting times:

• Visual Studio 2008 Professional ISO (3,47 GB) copied from a USB external disk: 4m34s
• Complete installation process of VS 2008 Pro: 9m47s
• Complete installation process of VS 2008 Pro: 14m48

It is important to note some things about those times: I included the time to copy the ISO from a USB 2.0 external disk to the SSD in order to note how significant the copy from a external device is related to the actual installation process; and it is also important to take into account that the MSDN install process takes a lot of CPU as file processing by the processor is done in a heavy way (so disk is not the bottleneck).

I also copied an ISO file of Office 2007 Professional (571 MB) from the USB 2.0 external disk in 25 seconds and the install process was performed in 3m22 seconds.

Someday I will review these times by comapring them to the times caused by using the original eSATA hard disk. Ah, all those tasks were made before I installed nVIDIA chipset drivers so times might be even slightly better (my fault! O:-) )

The tests strengthen my idea that it is very important to spend less money on the best CPU or memory chips when the operating system is accessing files or swap space more than often.

Installing a proper Win32 developing environment on Windows 7

I am playing a bit with Windows 7 Professional 64-bit. So far, the experience has been awesome and I absolutely recommend changing to this wonderful OS if you are a Windows user. I have tried to install Visual Studio 2008 this morning and also the Windows 7 SDK. In order to be able to use the latter making use of the former, VS 2008 needs to be updated with its Service Pack 1.

After downloading the installer for the VS 2008 SP1 (please note that the download link leads to the Spanish version), all I got were installation errors. In order to have it properly installed, all you have to do is execute the installer in Windows Vista SP2 compatibility mode.

About the Windows 7 SDK, you have to be aware of the two alternatives for a proper installation process: visit the link for the Web Installer (a small program that reads information about your Windows 7 system and downloads the files which it needs) or visit the page for the download of an image.

Also take into account that there are three ISOs for the three possible target CPU architectures (x86 for 32-bit CPUs, IA-64 for Intel Itanium chips and AMD64, which is the proper one for the Core 2 Duo in the laptop that I am using). If no comments show up in this post, it means that no install problems were found -I am still leeching the 1.5 GB of the SDK). Ah, I forgot one important thing: this is only for the underlying CPU architecture in your computer. Visual Studio allows you to generate code for all the three architectures, no matter which CPU you have.

I do not play with the SDK for pleasure. I want to test if I can compile both the Helix Server and Player for Windows 7 for the work of my PhD Thesis. If not, I guess I will need to install a new partition with Windows Vista or, more likely, Windows XP (where I have already done it in the past). Everytime that a new SDK comes up, headaches reach the Helix Community.

Google Latitude badge

Activated! You can check it on the left (this makes no sense if you read it via RSS... go to my blog) ;-)

Accessing an ext3fs filesystem from Windows Vista

I was a happy user of the kernel driver for Windows XP made available from fsdriver.org (the so called IFS driver). It allowed me to mount an ext2fs/ext3fs partition as a drive letter in Windows and treat it as any other disk.

After moving recently to Vista (not my fault, it was preinstalled in the new machine), IFS -which was supposedly meant to suppor the latest flavour of the Microsoft OS so far- asked me to reformat the drive in order to be able to use it. Gasp!

Googlin' here and there I managed to find the ext2fsd project in Sourceforge. I am trying it and so far it works fine. I only dared to mount the ext3fs partition in read-only mode by now and it is very likely that it will stay this way. I do not want to play with very serious things and most of my work is in that driver.

New Eee PC 901 in my hands

I recently bought an Asus Eee PC 901 in Redcoon. Great offer which made me buy it: 195 € plus 2 € and something (as I shared delivery fee with my colleague David).

If you follow my blog, you know that I played with the Acer Aspire One A150 that I gave to my wife last summer. I find the AAOne keyboard much more comfortable but I am getting used to the tinier keyboard of the Eee little by little. On the other hand, the new machine already has a 6-cell battery (I had to give a 6-cell battery to my wife for her AAOne as it barely lasted 1 hour and a half under Linpus Lite) and it lasts for over 6 hours in power saving mode while making use of WiFi and web browsing from time to time. I was at work so I could not make an intensive use of it.

The next test will be making the CPU suffer in High Performance Mode and see how autonomy decreases. Just for the sake of testing, as autonomy is a priority for me and Power Saving mode seems to be enough for browsing and editing simple docs.

One thing that I do not like is that the disk is divided and system partition goes easily full. For instance, I updated BIOS via Asus Updater and keyboard layout was lost. I launched kcontrol and changed keyboard layour from en-UK to es-ES and the keyboard model to A4Tech KBS-8. The terminal windows from which I launched kcontrol was indicating an error when changing keyboard layout. The reason was /dev/sda2 full so I gained some space by launching a sudo aptitude clean to remove cached .deb files.

I feel pretty comfortable with the default Xandros OS but partition layout might lead me to try other options as Easy Peasy, Windows or both O:-)

The Xandros OS is a bit better than AAOne's Linpus lite (at least for my taste). This machine also has a larger (16 GB vs 8 GB, plus the 4 GB internal flash) and a bit faster SSD disc, double amount of RAM (1 GB vs 512 MB), twice as much cells in the battery (plus a better energy management that leads to longer autonomy), same quality 1.3 Mpx web cam, Bluetooth support (gained in the AAOne thanks to a super tiny BT Dongle), and less size (with the inherent keyboard problems) but some more weight (due to the different size of the battery... they seem to equal when chaging AAOne's battery).

I'll keep reporting about my experiences with this machine.
So far I have fallen in love with it.

A new lesson of Real Sporting supporters

Another weekend for Spanish Leaghe BBVA and Real Sporting supporters, "the big red-and-white tide" will be in big number in Villarreal to support their team.

In these hard moments, "nunca irás solo".

Playing with the G1

In the 2008 W3C Plennary Week, Adam from Google was very kind letting me play with his HTC G1. It was my first experience with Android. The UI was pretty simple and easy to use although it lacked some interesting features like offering the on-screen keyboard when in landscape orientation.

This post is pending since october 2008 and it is now the moment after just receiving both pics from Jonathan Jeon (ETRI-Korea). Thanks a lot, Jonathan and Adam!

M.I.T. OpenCourseWare

By chance, I discovered this site where the people from the MIT leave for free some material of courses of different disciplines.

I reached the site after googling and finding the initial course on Classical Mechanics by the Department of Physics. There are lots of video material (and no excuse if you are not experienced in the English language, because there is a transcription for the voice in them -so you can use an automated translator or your fave English-speaking friend-).

Why not start from the first lecture about units, dimensions, measurements and associated uncertainties, dimensional analysis, and scaling arguments.

If you want Computer Sciences, they are also there.

Aren't AMOLED foldable displays lovely?

And promising, in the field of mobility. Few to say, just watch the prototype in the CES in Las Vegas these last days.

Animation about concepts to be aware of when designing mobile user experience

I encourage you to watch this animation as seen on piksels.com. Funny and interesting!


Designing for Mobiles: Users & Context from Niek Dekker on Vimeo.

W3C mobileOK Basic Tests 1.0 becomes a W3C Recommendation

No, I am not transcribing the W3C News feed, if you also read the last post from some minutes ago. mobileOK Basic Tests 1.0, the specification whose authors are Sean Owen (Google) and Jo Rabin (dotMobi -and before, at Segala-), has been demonstrated by means of a main reference implementation: the W3C mobileOK Checker.

Behind the web front end in that page, there is an open source checker that has been possible thanks to the mobileOK Checker Task Force: Abel Rionda and Miguel García (Fundación CTIC, current co-chairs of the TF), Sean Owen (Google, chair of the TF from its creation in early 2007 until June 2008), Jo Rabin and Ruadhan O'Donoghue (dotMobi), Dominique Hazäel-Masseux and François Daoust (W3C), Roland Gülle (7Val). Ah, and myself.

Basically, the open source checker implements the tests and the tests intend to provide algorithms for those best practices in the Mobile Web Best Practices 1.0 (Basic Guidelines) W3C Recommendation that can be performed by an algorithm without the need for a human expert (that is another battle that I'll write about someday).

This is part of the support of Fundación CTIC to web standards in what regards to mobility since September 2005, when I joined the W3C Mobile Web Best Practices working group:

• Provision of two implementation reports for the Mobile Web Best Practices W3C Rec:
  
  • "Implementation D" is the wonderful work made by Martín Álvarez Espinar on the web site of Fundamentos Web.
  
  
  • "Implementation E" is the official web site of the City Council of Zaragoza (zaragoza.mobi), whose content transformation engine was created from scratch by Luis Miguel González, Rodrigo García, Julio Argüello and myself. It was the "kernel" of Merkur, the content transformation proxy of Fundación CTIC.


• Implementation of a W3C mobileOK Basic Tests 1.0 checker: the subject of this post. By the way, in CTIC we have a great checker called TAW that can validate web sites against WCAG and mobileOK tests. Miguel García, co-chair of the mobileOK Checker Task Force, is the main analyst, designer and developer of this tool.


• Creation of the Device Description Simple API W3C Recommendation.

It has been a lot of work, besides the dozens of projects accomplished in the meantime, but it is good to see that results are visible and -what is more important- useful.

Device Description Repository Simple API is a W3C Recommendation

The DDR Simple API document has just become a W3C Recommendation. This work intends to provide a standard way for any stakeholder in the web ecosystem to access device information known "a priori" like screen size, number of colours, or image formats and markup supported. The document has been developed within the W3C Device Description Working Group and its editors are Jo Rabin (from dotMobi -mTLD Top Level Domain-), José Manuel Cantera Fonseca (from Telefónica I+D), Rotan Hanrahan (from MobileAware, chair of the DDWG) and myself, Nacho Marín, from Fundación CTIC.

It would be completely unfair not to mention the fifth horsemen, Mr. Rodrigo García (also from Fundación CTIC) who was a key player in the development of the document, the re-expression of the API in IDL and WDSL -to prove how it could be used by other technologies different to the Java API used in the document).

This is not the only document provided by the DDWG, so here goes my acknowledgment to: Andrea Trasatti (Director of Device Initiatives at dotMobi), as editor of the DDR Core Vocabulary (together with Jo and Rotan); Jose Cantera once again, as editor of the document Device Description Structures; Eman, James and Matt, as editors of the Device Description Landscape 1.0; and Rotan, yes, Rotan again, as editor of the Device Description Ecosystem 1.0.

And of course, to all the participants in the group: it was a pleasure to work with you all. I hope we'll work together again. Some of you are still doomed to bear with me in the W3C Mobile Web Best Practices Working Group, one of the other groups in the W3C Mobile Web Initiative.