Standards
Posted January 19th, 2004 @ 09:00am by Erik J. Barzeski
Dave Hyatt talks about coding to standards accepting malformed content. I, like David, want want a browser that doesn't accept crap, or even a browser with a "No Crap" mode. Unfortunately, such a browser would limit my browsing to this blog and the FSS site (and even they wouldn't be available sometimes).
iCab used to have a smiley face that rarely, if ever, smiled. iCab failed as a validator because it never really supported CSS. Now it doesn't support XHTML, remaining useless. Omni Group had a huge opportunity, in hindsight, to add validation features to OmniWeb 5.0. I'd pay (again) for those! Instead, based on the OmniWeb 5.0 previews, it seems Web developers will have to satisfy ourselves by creating bookmarklets to http://validator.w3.org/check?uri=?
.
How could Omni Group have earned another $30 from me (and a whole heap of thanks)? Easy. They could have added a drawer to the main window which, when open, showed validation results. I might leave such a drawer open full-time out of curiosity! With this Validation Drawer, I'd have one window with both content and validation results - form and function, appearance and compatibility. The Validation Drawer would mean that Web designers on the Mac would have little excuse for creating malformed pages.
Currently, validation is an exercise for the non-lazy. Validation requires a lot of copy/paste, at least one more window, and a lot of reloading (of two windows). That may not sound like much work in the end, but it can feel like a daunting task when you have 50 pages to validate because you haven't done it in awhile. How often do developers visit their own sites? Fairly often. How often do they validate? Almost never. A Validation Drawer which remained open would narrow that ratio to 1:1 - they'd be getting validation (or alerts for non-conformity) every time they visited.
I just spent a morning quickly making sure all FSS URLs validate. Here are the results, and a *
= "passed." Only two pages really needed tweaking, and my CSS file has validated since its inception:
* / * /products/ * /products/maildrop/ * /products/maildrop/features.fss * /products/maildrop/versionhistory.fss * /products/maildrop/screenshots.fss * /products/ftpeel/ * /products/ftpeel/features.fss * /products/ftpeel/versionhistory.fss * /products/ftpeel/screenshots.fss * /products/pulpfiction/ * /products/rockstar/ * /products/iwipe/ * /products/booklet/ * /products/ichem/ * /products/freeware/ * /support/ * /support/maildrop/ * /support/ftpeel/ * /support/pulpfiction/ * /support/rockstar/ * /support/other/ * /aboutus/ * /aboutus/privacy/ * /aboutus/geeksforhire/ * /aboutus/mailinglist/ * /aboutus/technology/ * /contactus/
I keep this list up-to-date and validate each page after I've made changes. Tedious, tedious, tedious. I've created an AppleScript that opens the URLs in Safari tabs with the validation URL prepended, but again, this is tedious.
I'll try to keep my hopes up for OmniWeb 5.1. My feet are planted firmly on the soil of standards. Unfortunately for everyone, many will never make it to shore.
Posted 19 Jan 2004 at 10:43am #
Web Standards
Erik mentions an excellent feature that web browsers should have in his NSLog(); - Standards entry. A window/drawer that you...
Posted 19 Jan 2004 at 10:52am #
Serving XHTML with the application/xhtml+xml MIME type gives those of us on Mozilla-based browsers that functionality right out of the box, though the practice itself is fraught with perils.
Worth looking into, though. The list of X-Philes over at Evan's place continues to grow.
Posted 19 Jan 2004 at 11:39am #
I have often wanted wanted omniweb to have that validation feature as well and a better bookmarking system (at least we got a damn good one at that)
Posted 19 Jan 2004 at 4:10pm #
I'm with you 100%. The only page that doesn't validate on my site is one with a quicktime movie in an embed tag. I could probably apply the Satay method to fix it, but haven't got around to it yet.
On a side note, for those with BBEdit, Markup > Check > Site Syntax will fly through your pages validating all the HTML/XHTML and even CSS (I think).
There's always the W3C Markup Validation Service too.
Posted 20 Jan 2004 at 1:23am #
As Chris mentioned, there's the W3C validation service. I use that. Of course, it sounds like you're trying to validate it on your computer instead of putting it live and then validating.
On a happy note, I think I might actually have completely ridded my new site of dead links and invalid markups and the like. Of course, odds are that I missed something, and it's still got a ways to go before I stop messing with it, but it's good for now :).
Posted 20 Jan 2004 at 2:13am #
I think I remember hearing about bbedit doing that but it really does work too well when your pages generate on the fly using a database you need the browser to check the output. maybe there are licensing problems with putting the validator in a browser.
Posted 20 Jan 2004 at 8:08am #
Jo-Pete, no, I'm not trying to validate it on my computer, and both you and Chris should re-read my post: I mentioned the W3C validation service too. C'mon now…
Plus, as Jonathan points out, BBEdit kinda fails if you do things like:
<? include('head.inc'); ?>
In other words, your document type declaration appears in only one file. The only way to test your page is to do it live, with files included and database output, uhhh, outputted. 🙂
Posted 21 Jan 2004 at 10:02am #
Today I received an email which I've summarized for you in five words: Your Site Sucks, Hire Me. Perhaps I am feeling particularly {bored, mean-spirited,...
Posted 27 Jan 2004 at 8:05pm #
Greg, of the Omni Group, is talking about OmniWeb 5 in a forum: 2) It is true that scrolling vertically is a lot more "comfortable" than scrolling horizontally, so while horizontal real estate is generally more available, you do want...