malwk (malwkgad) wrote,

End of an era or how we should not favour our past on the expense of the present day

Dear Mr. Mandla,

I am happy to report that my adventurous love affair with your blog is now over.

I find it mostly waste of time for the following reasons:

  • while all your arguments and howtos are viable and practical, you tend to proclaim the fallacy that:
    using console interface is the path to the true computing, which is (scientifically and measurably) untrue
    Using keyboard only to navigate between several contexts might look to you like you are more effective, but it is actually not, it just seems that way for you. Adequately measuring the performance from a remote stand point will reveal that this is not the truth, it just is apprehended that way by the operator/user because the brain is actively involved in doing the switching between the contexts, while when using the X (or any other window environment) this is mostly 'brainless' operation, your hand moves the mouse to the point where your eyes see the task/context, you proceed to the task as you see it, you don't have to know where it is, it is simple mechanical exercise.

  • your statement discards the true nature of the www
    while it is true that some web authors tens to overuse the styling capabilities of the modern browser environment, it is also entirely true that the user have complete absolute control over the presentation of the information in the browser
    css can be disabled completely, image loading can be disabled, javascript can be disabled, advertising can be disabled, plugins can be disabled. More over, the learning curve for text only browser compared to one time configuration of a modern browser to display the information as text only browser is significantly more acceptable. One can share its browser configuration and while text only browser configuration can also be shared between users still the high learning curve for navigation and adequate advanced uses remains, while regular browser learning curve is pretty much one step - learn to use the mouse to point to an object and click. More over - it requires expertise way beyond the one of the average user to configure video/picture display (content that IS important, unless one is only interested in technical documentation, because all other documentation can benefit from diagrams and other image representation of the data) which is point 3 in your fallacy statements

  • all image/video/graphics/sounds embedded in a web page are useless and distract the user from the actual content
    while this can be true in some cases, most viable sources of information at least try to use the tools of modern day web environment for the benefit of the user and not for distraction
    Again, the user have complete control over the user experience received from the web (which is NOT the case for regular application, will get there in a minute). Lots and lots of web applications simply would not exists without the modern capabilities of the browser environment. Applications that ARE useful, like web calendars (I will not go into the cloud vs personal data here, you can run your calendar application on your server as well as on google/yahoo and still make it available on your phone!), banking applications, data mining and visualization, real time data presentation and so on. Using audio directly in the browser is very scarcely used anyway, while using video to present the outlines of a big chunk of information or demoing a large project is very popular and NOT because people are lazy to read the full article, it is just more easily apprehended because it involves presenting more dense bits in shorter time. It is as simple as that, sorry but it is truth. Also images ARE important, how fast can one describe the snow on his street in words and in picture? Even you have to admit it. And it is also true, today 99% of the pictures are exchanged via http. Take for example the density of information you can get from a photo blog using elinks and firefox/webkit - load the blog and go... I bet I will be finished with the blog before you can even load the first 10 pictures. Talking about effective use of our time, not "right way to use computers"

  • console applications are more useful: i don't even have to comment on this one, but I will
    while there are some very capable console application, (for example newsbeuter is extremely powerful in its filtering!), the learning curve for them and the usability are much lower, which makes the statement that computing is not for grandpas true, which thank god is not anymore!
    First of all pointing and clicking on well known icon / signs (like the play sign which is UNIVERSAL and well known and easy to remember and recognize) is much much more easy to learn and easy to use than remember how the same action is performed in each and every application (as console applications tend to NOT follow some sort of standard on how things are done, there are almost no music/video players to make the same thing the same way in console world). Second - contest switching from point one is valid here too - one does not really have to switch context first before perform action in another context, while 2 or more contexts can be observed and acted upon if desired.

    Lets say I have browser and torrent client. In console I have to first switch from browsing to torrent context, then perform the desired action. In regular GUI i just perform the action should both context are observable (which (if they are) I can decide and not rely on specific predefined configuration like I have to in console (using screen or similar utility)).

    And then we have again density of information: representing actions as buttons with well known signs / icons is very powerful, as the sign is the same while the action can be relevant: example: in torrent client play sign will start the file while in music player the same will play the song, in IDE it will run make, in service configuration tool it will start the service. The user can guess the action from the context of it, while in console application while some provide "help" menu somewhere the user mostly HAVE to know how to perform an action. eventually most user will learn, but it is pointless and wasting time. The performance in account of the user interaction will never reach the one of GUI app.

    Here we face also the language barrier: most applications are written in English, should the user NOT know the language - we are done. Some console apps can be translated easily, some might not at all, but using icons and buttons IS universal. So stating that using console applications is more powerful is complete fallacy, how powerful an application is depends on the developer not the medium. The same applies to its resource demands: while it is true that pretty much all GUI apps require more resources to run, it is also true that this depends entirely on the developer: I for one have developed WEB interface for an application and the GUI for it is run on the browser, while the application work is headless. The application of its own requires as low resources as one written for the console and still provides more useful interface to its functionality. The fact that most applications rely on GTK/QT/KDElibs is true, but it is related to laziness and speed of development, the original Xlib was a small hell and very very hard to use. This is why GUI libs are predominant, they open the computing world to more developers. Same with Web apps - making useful things with javascript is so easy, pretty much anyone can start using it!

    Web app can be very easily translated and can be upgraded and updated as much as the author wants, edge cases (where for example no JS is avalable) can also be handled, more over the application can be easily modified on the user side (using client scripting for example - something that is mostly impossible for you for example, as you do not know C and cannot modify the applications you use, you rely entirely on the developer).

I am not sure if anyone here will agree on any of this, but now I am completely sure you and your followers apprehend yourselves as some cast that knows the right path to the computing and everyone else for you is lost in the non-sense of modern computing. I was trying to follow your path of thinking and made this experiment, I have lived without normal computing environment for over a month. It slowed me down, It made me spend more time in front of the computer and it made me lag on performing very simple tasks (for example copying text from one application to another is pain in the console when the application has its own windowing, like finch of irsii or elinks!!!).

This said - now I believe you are on the wrong path, for users like you I believe the right path would be - run web applications (your own!), write them to be fast and efficient and have only web interface, very simple one, but web, and use whatever you feel like as browser to access all of them! browser environment can be effective even without images, sounds and plugins! one can also configure handlers for flash video in a regular browser ( so the video can be played by mplayer or just downloaded and so on), all images can be saved at once and on and on possibilities, that you delude users into forgetting and hang themselves in an era not 10 =12 years ago but 30! because text terminals were used 30 years ago! Just watch Crockford's first hour presentation on history of computing and you'll see what I am referring!
Tags: gnomedesktop, thoughts
  • Post a new comment


    default userpic

    Your reply will be screened

    Your IP address will be recorded 

    When you submit the form an invisible reCAPTCHA check will be performed.
    You must follow the Privacy Policy and Google Terms of use.