in reply to Re^4: Better perldoc on Windows
in thread Better perldoc on Windows

I used to use Lynx a lot when I had dial up--for its speed which came mostly from the fact that it didn't download images and flash crap etc. Then I discovered Opera that allowed me to defer image download until I chose to download them--individually or on-mass, and I didn't use Lynx much after that.

These days, now I have broadband, speed is less of an issue. And with css and javascript being de rigour, it would be pretty much useless anyway for online browsing. For local html, download speed was never an issue, but the restrictions of fix-width only fonts, no css or javascript is. I really cannot see the benefits of the intermediate stage. My browser is always there, and I can quickly flip between several ModPods and my editor and one or more command lines quickly and easily.

Trying to do everything through a single UI makes no sense to me. Like trying to control one of the modern breed of photo-realistic car or plane sims through a keyboard: possible, but crappy. Can you imagine trying to drive your real car by pounding on the uparrow to go faster or downarrow to hit the brakes?

For all it's limitations, html is far superior--for users--to any of the other text viewers, simply because its ubiquity, and open competition, has meant browsers have developed rapidly to improve the UI.

Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.
"Science is about questioning the status quo. Questioning authority".
In the absence of evidence, opinion is indistinguishable from prejudice.
RIP an inspiration; A true Folk's Guy

Replies are listed 'Best First'.
Re^6: Better perldoc on Windows
by LanX (Saint) on Jun 07, 2010 at 02:51 UTC
    > I really cannot see the benefits of the intermediate stage.

    a) This thread is about local POD from perldoc, so no CSS, no JS, no speed problems, no images and (sorry) even no modern plane simulation.

    b) w3m does not only allow to execute shell commands directly (type !) it also supports many keybindings of your preferred ed(v)itor. No mouse needed. (Sometimes you just wanna automatically search for a special keyword or only display a certain paragraph)

    c) Using w3m within an emacs buffer¹ allows me running the same automization interface (macros) for POD and my source, like cutting code, switching buffer and pasting and running it, without even touching the mouse. The geometry of splitted windows showing my source and POD are automatically controlled, the bookmarks use the same interface, and all the other "same UI benefits" I forgot...

    d) And as I said, it's only optional, users could still use a full scale browser if they don't like these benefits...TIMTOWTDI...

    (Sure I could emulate all of this with JS, AJAX, a local podserver or sending FF remote commands and FF AddOns like vimperator and It's all text, plus tools like AutoIt or wmctrl but surely for more than doubled implementation costs)

    Cheers Rolf

    UPDATES: 1) added link

[reply]
      ) This thread is about local POD, so no CSS, no JS, no speed problems, no images

      Hm.

      1. No CSS: 
        <html>
<!-- saved from url=(0017)http://localhost/ -->
<head>
  <title>ActivePerl User Guide</title>
  <meta name="Author" content="ActiveState">
  <meta name="AppleTitle" content="ActivePerl 5.10.1.1007 Help">
  <script language="JavaScript" src="tocTab.js"></script>
  <script language="JavaScript" src="tocParas.js"></script>
  <script language="JavaScript" src="displayToc.js"></script>
  <link rel="stylesheet" type="text/css" href="Active.css">

        [download]

        And very useful it is too. To be able to tailor *all* the POD on my set up to my preferences--fonts; sizes; colors--with simple changes in *one* place.

      2. No JS:

        Wouldn' t it be nice to be able to (say) double-click (or Ctrl-Alt-Meta-Meta-f) a function/method name in the POD and be taken directly to the module source for that function/method?

      3. No images:

        And no amount of words will ever be a substitute for a few simple pictures.

      I guess the only lesson here is that you hate my way of working as much as I hate yours.

      Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.
      "Science is about questioning the status quo. Questioning authority".
      In the absence of evidence, opinion is indistinguishable from prejudice.
[reply]
[d/l]
[select]
[reply]
        > I guess the only lesson here is that you hate my way of working as much as I hate yours.

        Nonsense, I always said TIMTOWTDI, but what I certainly dislike are dogmatic approaches and polemical tricks to defend them...

        ...so - as usual - lets better plonk this conversation. :)

        Cheers Rolf

[reply]
[reply]
      Using w3m within an emacs buffer¹ allows me ...

      No need to install w3m, WoMan is core in emacs.

      cperl-mode also has an info-reader interface to perldocs!

[reply]
        > No need to install w3m, WoMan is core in emacs.

        sure that's what I meant with "I'm never running into these kinds of problems".

        > cperl-mode also has an info-reader interface to perldocs!

        Well theoretically... Unfortunately this relies on installing static info-files (for CORE modules only) which stopped being updated with Perl 5.6 (IIRC).

        While it should be fairly simple realizing a dynamic pod2info translation and displaying in emacs and I'm a big fan of Info when it comes to navigate large docs as a tree!

        But I don't really see the benefit over porting the info semantics and keybindings to (wo)man or html display, since well maintained perl packages are already splitted into different modules... So using a podserver has much more benefits...

        Cheers Rolf

[reply]

In Section Cool Uses for Perl