Beefy Boxes and Bandwidth Generously Provided by pair Networks
laziness, impatience, and hubris
 
PerlMonks  

Quality Standards Proposal to HTML Developers?

by metaperl (Curate)
on Nov 02, 2005 at 19:26 UTC ( #505066=perlmeditation: print w/ replies, xml ) Need Help??

Our group is keying up for some serious dynamic HTML generation. I want to raise a high bar to the HTML group on the quality of HTML they can present to me. I would appreciate some feedback on what you do as a developer to insure that you get HTML easy to modify and understand. Here are the standards that I want to propose to our HTML team:

Dear HTML Group, in order to make sure that I get quality HTML from you, I am requesting that you do the following things before submitting HTML to me:

  1. Validate your HTML using the W3C Markup Validation Service or the WDG HTML Validator

    If there are any errors or warnings, correct them, or provide adequate justification for not fixing them.

  2. Make sure that the final HTML delivered is line-wrapped to 120 columns and indented for readability using the tool of your choice. I would point them to HTML Tidy but the up to date versions only run on Linux

Comment on Quality Standards Proposal to HTML Developers?
Re: Quality Standards Proposal to HTML Developers?
by jeffa (Chancellor) on Nov 02, 2005 at 19:48 UTC

    Is it possible for you to build a CGI app (running on a Linux box) that shoves the HTML through HTML::Tidy for your developers?

    jeffa

    L-LL-L--L-LL-L--L-LL-L--
    -R--R-RR-R--R-RR-R--R-RR
    B--B--B--B--B--B--B--B--
    H---H---H---H---H---H---
    (the triplet paradiddle with high-hat)
    
Re: Quality Standards Proposal to HTML Developers?
by johnnywang (Priest) on Nov 02, 2005 at 19:55 UTC
    From my experience, the most important things for HTML are:
    1. use CSS, never hardcode styles
    2. a correct template system: include dynamic content in a static layout, rather than include static content (e.g. header) in dynamic pages. Both TT and Mason supports this.
      I agree, completely.

      Our single most important thing is CSS and templating.
      Er, our *two* most important things are CSS, templating, and strict guidelines on page-load speed.
      Wait, I'll come in again. *Amongst* our most important things are....

      -- WARNING: You are logged into reality as root.
Re: Quality Standards Proposal to HTML Developers?
by philcrow (Priest) on Nov 02, 2005 at 19:56 UTC
    I might change the tone slightly by revising:
    ...to make sure that I get quality HTML from you,...
    to the more neutral:
    ...to make sure that we deliver quality HTML to our customers, I am requesting that you do the following before submitting HTML:...

    If I were in the HTML group that wouldn't raise my defenses quite as fast or as high.

    Phil

Re: Quality Standards Proposal to HTML Developers?
by derby (Abbot) on Nov 02, 2005 at 20:02 UTC

    I'm not following, is the HTML group a design group or a development group? If it's design, then #1 and #2 are good.

    There may be some other things to ask for depened upon who the consumer of the HTML will be. Think of javascript rules (yes, no - to what model), graphics rules (little, lot, provide psd files not just flattened gifs or jpegs), and CSS rules (valid, hacks well documented, selectors well named).

    I think the use of validators is great but agree on the one(s) you'll use up front (the last project I worked on, the agreed upon validator was "looks good in IE" - ug).

    -derby
      I'm not following, is the HTML group a design group or a development group?
      design.
Re: Quality Standards Proposal to HTML Developers?
by ickyb0d (Monk) on Nov 02, 2005 at 20:21 UTC

    You might want to also specify which W3C Specifications that you're looking to meet (XHTML 1.0 strict, HTML 4.01, HTML 4.01 Transitional, etc.). The same goes for CSS (1.0, 2.0).

    As stated above, you might want to consider your customer/client. Should this webpage be geared towards IE, Mozilla based browsers, Opera or all of them. There can be completely different 'quality' versions of HTML based simply on browser type.

Re: Quality Standards Proposal to HTML Developers?
by Anonymous Monk on Nov 03, 2005 at 12:57 UTC

    The current Windows version of Tidy is good enough.

    Scavenge anything you need from NYPL StyleGuide

Re: Quality Standards Proposal to HTML Developers?
by samizdat (Vicar) on Nov 03, 2005 at 14:17 UTC
    Say 'please', as someone else suggested

    Choose which HTML/DTD version. 4.0 is pretty stable.

    Validate it, as you said

    Decide upon a browser target. IE and Moz are very different in rendering fonts and tables

    Use CSS 1 and JS 1.2 only

    I also would push for a templating system and separate out the logic. I have used HTML::Embperl, a custom PHP4 one, J2EE and RoR, but YMMV.

      If I could only downvote you enough for suggesting that the site code to only one browser ...


      The intelligent reader will judge for himself. Without examining the facts fully and fairly, there is no way of knowing whether vox populi is really vox dei, or merely vox asinorum. — Cyrus H. Gordon
        idsfa, I'm sorry, but reality sucks. I used to be as adamant as you are on this very subject, esp as I was trying to debug sites using my FreeBSD desktop.

        Then I tried to make a site that was used by marketdroids and HR people. 98.9% use Doze and usually the latest.

        Do I like it? NOT.

        We are making the attempt with our latest version. We are rewriting to enable best possible low level compatibility but even Google can't make a simple webmail cross-platform compatible with all their $$$ and J2EE. We still have a whole bunch of logic traps in there for various JS and browser incompatibilities, and spend more time debugging that than writing business logic. Being browser neutral costs, idsfa. It's hard to justify when your clients are all on one platform.
Re: Quality Standards Proposal to HTML Developers?
by Juerd (Abbot) on Nov 04, 2005 at 11:59 UTC

    line-wrapped to 120 columns

    Why 120? That's not a very common size. Pick 80 instead. This is, of the screens with a fixed size, by far the most used. If the HTML is clean, 80 columns will not be a problem, if the indent is small enough. I prefer 2 spaces for HTML, 4 for code.

    This said, I think one thing is severely wrong with your post. Quality is FAR more than just style and correctness.

    There are some important issues you say nothing about. Pick a scheme for consistent class/id naming, and stick to it! Begin all script variables with 'x', to avoid clashes with elements in MSIE! Make sure everything works in Gecko! Define a scripting style! Etcetera! :)

    Juerd # { site => 'juerd.nl', plp_site => 'plp.juerd.nl', do_not_use => 'spamtrap' }

Re: Quality Standards Proposal to HTML Developers?
by CharlesClarkson (Curate) on Nov 06, 2005 at 18:36 UTC

    I think this places too much of a limit on the designers. I like xhtml, but somtimes I used HTML 4. Well formed mark up is great, but I might want to use IEs -1 value for tabindex in forms. That won't validate under any HTML 4.01 versions.

    Your designers will love templates. They can play with new designs without bothering you guys while still using curerent data. And your code looks much cleaner when using templates than when using embedded markup.

    HTH,
    Charles
Re: Quality Standards Proposal to HTML Developers?
by delegatrix (Scribe) on Nov 09, 2005 at 16:54 UTC
    Have you company institute web development standards, at the very least for which markup to use and that it must validate. With standards, you don't have to 'suggest' that anyone 'please' do anything. You can just say 'Your stuff is not up to the standard. It needs to be.'

Log In?
Username:
Password:

What's my password?
Create A New User
Node Status?
node history
Node Type: perlmeditation [id://505066]
Approved by samtregar
help
Chatterbox?
and the web crawler heard nothing...

How do I use this? | Other CB clients
Other Users?
Others rifling through the Monastery: (4)
As of 2014-07-26 05:30 GMT
Sections?
Information?
Find Nodes?
Leftovers?
    Voting Booth?

    My favorite superfluous repetitious redundant duplicative phrase is:









    Results (175 votes), past polls