Programming Forums > Web Development and Design > HTML / XHTML / CSS

ExplainThat · Oct 18th, 2007, 11:13 AM

Quote:

Originally Posted by cscgal

As far as I am concerned, every single tag that's opened must be closed. If there is no reason for a tag to be closed (such as an image tag or line break) then it should be self closing by ending with a slash as in <tag ... />. Just because it doesn't error out doesn't mean it's okay to do. I feel a C++ void main() lecture coming on ...

---

What often happens is web browsers will implicitely add a closing tag when it is impossible for it to remain open. It's impossible to remain in the <head> section once the <body> has started, so let's implicitely close the head tag before the body. This same thing happens with lists. A lot of people type <li> at the start of a list element without closing it as so:

(Toggle Plain Text)

<ul>
  <li>One.
  <li>Two.
</ul>

This is wrong, but it's impossible for something to be a part of two list elements of the exact same list, so a web browser will implicitely close the first list element before moving on to the second one. If you get into this habit, you'll inevitably not close something that technically could remain open (and therefore the browser won't be nice enough to close it for you), even though it's incorrectly nested or not what you intended.

I agree wholeheartedly - I was just suggesting a way to check on when it might be safe to leave out a closing tag for those who wish to do that. The reasoning being that if Google can get away with it chances are it works in most browsers.

cscgal · Oct 18th, 2007, 11:16 AM

Google can get away with a lot of things that us commonfolk can't.

DaWei · Oct 18th, 2007, 3:01 PM

Let me relate this issue to another of the same variety. Many compilers will accept "void main ...". Some compilers will also accept "fflush (stdin);". Others will accept "rewind (stdin);".

These are proprietary solutions that detect the difference between input buffers and output buffers and deal with them accordingly. Any reasonably proficient programmer, given a knowledge of the underlying implementation, can achieve solutions for these "problems".

C and C++ chose not to address, directly, implementation issues. I personally believe that this is the correct approach. There's little doubt that this personal opinion is connected to the fact that I have built many systems with many kinds of keyboards and displays, and even systems without (except in test/debug modes) keyboards and displays, at all.

The bottom line is that performance, however widespread, does not dictate the standards. Standards define compliance. Anything that does not meet the standards is non-compliant, unless the standard explicitly denotes freedom in the area, by deliberately not defining or not specifying some particular response.

cscgal · Oct 18th, 2007, 4:40 PM

I was right! I predicted on my post on page one: I feel a C++ void main() lecture coming on ...

WaltP · Oct 18th, 2007, 6:48 PM

Quote:

Originally Posted by cscgal

I was right! I predicted on my post on page one: I feel a C++ void main() lecture coming on ...

Gee, she's psychoti... er... psychic!

joeprogrammer · Oct 18th, 2007, 10:45 PM

Quote:

Originally Posted by ExplainThat

I tend not to leave out closing tags even when I know that it works. However, on the rare occassion that I have to I check to see what Google is doing - just take a look at their SERPs. There are tons of missing closing tags - for instance, they hardly ever close a listitem. The way I figure, if Google is skipping a closing tag then it should generally be safe with most browsers since I cannot imagine that they would leave any browser that could possibly matter out in the cold.

MSDN has a lot of code examples that use void main. Does that mean it's okay to follow Microsoft's example, too?

Personally, I agree with Dani. The stricter you conform to the standards, the less likely you're going to screw up, and the less likely your website is going to screw up on some crap browser.

Jimbo · Oct 19th, 2007, 1:12 AM

Standards compliance is also becoming a Good Thang(TM) in the industry as well, so we'll likely see many improvements in that so that there's less issue about those "some crap browsers" out there.

funcravers · Dec 26th, 2007, 1:47 PM

It's a bad habit not to close any tag. The only tag I don't close is the <HTML> tag because I always forget when I'm finished with editing a webpage.

Kuryn · Dec 26th, 2007, 5:55 PM

Well, at least Microsoft is making up for it in Internet Explorer 8 since it did pass the Acid2 test, even though it don't really matter. All the Acid2 test passing for developers. Not like it will matter to the non-developers.

And it sucks that people have to have all CAP tags which is badly designed, or optional "" in the attributes, to optionally omit end tags.

cscgal · Dec 26th, 2007, 10:45 PM

Quote:

Originally Posted by funcravers

It's a bad habit not to close any tag. The only tag I don't close is the <HTML> tag because I always forget when I'm finished with editing a webpage.

Isn't that like the worst tag to forget?

Oct 18th, 2007, 11:16 AM	#12
cscgal Hobbyist Programmer Join Date: Oct 2007 Location: New York Posts: 163 Rep Power: 10	Re: Optional Closing Tags in HTML Google can get away with a lot of things that us commonfolk can't. __________________ DaniWeb IT Discussion Community Everything Information Technology related all in one convenient interactive online publication and community

Oct 18th, 2007, 3:01 PM	#13
DaWei Resident Grouch Join Date: Jun 2005 Posts: 6,453 Rep Power: 10	Re: Optional Closing Tags in HTML Let me relate this issue to another of the same variety. Many compilers will accept "void main ...". Some compilers will also accept "fflush (stdin);". Others will accept "rewind (stdin);". These are proprietary solutions that detect the difference between input buffers and output buffers and deal with them accordingly. Any reasonably proficient programmer, given a knowledge of the underlying implementation, can achieve solutions for these "problems". C and C++ chose not to address, directly, implementation issues. I personally believe that this is the correct approach. There's little doubt that this personal opinion is connected to the fact that I have built many systems with many kinds of keyboards and displays, and even systems without (except in test/debug modes) keyboards and displays, at all. The bottom line is that performance, however widespread, does not dictate the standards. Standards define compliance. Anything that does not meet the standards is non-compliant, unless the standard explicitly denotes freedom in the area, by deliberately not defining or not specifying some particular response. __________________ Abstraction doesn't make it impossible to write bad code; it makes it possible to write superior code. Contributor's Corner: Grumpy on C++ Exceptions DaWei on Pointers

Oct 18th, 2007, 4:40 PM	#14
cscgal Hobbyist Programmer Join Date: Oct 2007 Location: New York Posts: 163 Rep Power: 10	Re: Optional Closing Tags in HTML I was right! I predicted on my post on page one: I feel a C++ void main() lecture coming on ... __________________ DaniWeb IT Discussion Community Everything Information Technology related all in one convenient interactive online publication and community

Oct 19th, 2007, 1:12 AM	#17
Jimbo Battle Programmer Join Date: Feb 2006 Location: Bellevue, WA, USA Posts: 754 Rep Power: 3	Re: Optional Closing Tags in HTML Standards compliance is also becoming a Good Thang(TM) in the industry as well, so we'll likely see many improvements in that so that there's less issue about those "some crap browsers" out there. __________________ <insert disclaimer here> <insert shameless plug for Visual Studio here>

Dec 26th, 2007, 1:47 PM	#18
funcravers Newbie Join Date: Dec 2007 Posts: 5 Rep Power: 0	Re: Optional Closing Tags in HTML It's a bad habit not to close any tag. The only tag I don't close is the <HTML> tag because I always forget when I'm finished with editing a webpage. __________________ ME Cave Community Profile\|\|Myspace Profile Cool Websites Cave Community\|\|Will Host For Food

Dec 26th, 2007, 5:55 PM	#19
Kuryn Newbie Join Date: Dec 2007 Posts: 14 Rep Power: 0	Re: Optional Closing Tags in HTML Well, at least Microsoft is making up for it in Internet Explorer 8 since it did pass the Acid2 test, even though it don't really matter. All the Acid2 test passing for developers. Not like it will matter to the non-developers. And it sucks that people have to have all CAP tags which is badly designed, or optional "" in the attributes, to optionally omit end tags.

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
start Learning HTMl	smita	HTML / XHTML / CSS	0	Mar 15th, 2007 7:38 AM
converting string to float	beginnerCCC	C	22	Oct 2nd, 2006 11:59 PM
Basics of HTML Tutorial	crawforddavid2006	HTML / XHTML / CSS	14	Jul 11th, 2006 7:40 AM
Basic HTML Tutorial - Reuben Keeney	ReubenK	HTML / XHTML / CSS	14	Mar 26th, 2006 5:50 AM
HTML =/= Programming	Sane	Coder's Corner Lounge	32	Mar 8th, 2006 2:50 AM

Programming Forums > Web Development and Design > HTML / XHTML / CSS

Optional Closing Tags in HTML