quoteless html

[Natko]

Hi,
some says that are mandatory, some says optional.
I have my html pages quoteless in function of optimization and works fine in IE5.5. Unfortunately, I haven't tested them on other interpreters, yet.
other expirience in this?
:-)

[rogerd]

Natco, you need to use quotes if you want your code to validate. Also, if your attribute contains a space, you'll need quotes. However, your code will probably work in most common browsers.

[TGecho]

As said before, you need to be careful when omitting quotes. Frankly, a better approach to optimizing is the write clean code in the first places. In other words, look into ways of using CSS to format your text, and possibly even your entire site. If done well, just getting rid of font tags can make a huge difference.

[Natko]

thanks, I know all that. I use external css and js, classes with only 1 caracter etc. I omit all defaults (cellpad=1 cellspac=2 table border=0), align over tr etc. Yes, you can't omit them for a 'space caracter'.
The only thing I'm curious is quotes. I have omitted them at all (in href too, but not in js) and it works fine in IE5.5
I wonder what are your experience in omitted quotes in html, and in js too. And if you have any experience in cross-browser NON-compatibility.

[tedster]

I put some pages on this kind of diet a few years back. There were just a few spots I ran into rendering trouble on NN4 and most of those were fixed when I added in those optional </p> or </li> tags.

I do recall that I had one (just one) Netscape 4 glitch that I could only fix by putting back one set of quotes. Sorry that I can't recall the specific details on that one.

[bufferzone]

A SE is like an old browser, not like EI, but just an old standard browser. Therefore you should always try to code your pages as close to standard as possible. Validate your HTML and follow the recommendations if you can. Quotes will be part of those recommendations.

IE is very forgiving and will tolerate a lot of errors and slobby coding. You might not find the SE’s as forgiving and I wouldn’t be surprised at all, if some SE’s take slobby coding as a sign of unprofessionallism and downgrade your ranking as a result.

[ergophobe]

I wouldn’t be surprised at all, if some SE’s take slobby coding as a sign of unprofessionallism and downgrade your ranking as a result.

FUD (Fear, Uncertainty, Doubt).

I can't believe that an SE would choke because of missing quotes. If their parsing were that rudimentary, they would just be looking for the end of the tag and wouldn't know about attributes at all.

The only reason SEs would take into account the "professionalism" of a site would be because the SE's designers believed that W3C goupies (which I am, so don't get riled) had a monopoly on useful content. In fact, I would say the people most likely to have real expertise on a given non-computing topic are least likely to be W3C groupies. They are hobbyists, fanatics, academics, archivists, consultants etc etc, and don't have the budget to hire someone unless they are selling something (and therefore are not primarily informational sites). Spelling errors might hurt because you won't get indexed for that term, but the poster is not asking about sloppiness, but systematic and careful non-observance of a specific recommendation.

Personally, I always try to validate my pages to XHTML and therefore use quotes (though I do try to get rid of tab/space indentation, unnecessary whitespace and so on).

That said, I don't think there is a browser out there that will have a problem unless you have unmatched quotes (obviously) or you serve up your pages with an xml mime type. Then you're hosed.

The brief version of this rant is
- you're probably fine
- I would spend the extra 200bytes to have a validating site.

Tom

[bufferzone]

Ergophobe>I agree completely. Your point is exactly what I was trying to say, you just say it more elegantly then I do.

[Natko]

Well, I will leave it quoteless to see what happen on the next dance. To quote or to light, I hope that SEs will premiate 'light', Btw, I validated my pages with CSEpro_v6, and it says "your pages are OK, but to light, add some quotes"
:-)

[g1smd]

You must quote all "#FFFFFF" colours, all "50%" sizes, all URLs, any attribute containing a / character (meta content-type etc), and any attribute with a space in the value, as a start.

I prefer to "quote all attributes".

[Natko]

My metas are quoted. I use absolute sizes and no relative. All alt, title, and names with spaces are quoted. but color .... why you think it should be quoted? Becouse of interpreter or SE?

[grahamstewart]

w3c says:

In certain cases, authors may specify the value of an attribute without any quotation marks. The attribute value may only contain letters (a-z and A-Z), digits (0-9), hyphens (ASCII decimal 45), periods (ASCII decimal 46), underscores (ASCII decimal 95), and colons (ASCII decimal 58). We recommend using quotation marks even when it is possible to eliminate them.

[w3.org...]

[g1smd]

It needs to be quoted because the # is not on the list of attributes that can have the quotes left out.

Anything with # or / or a space in it must be "quoted".

[richlowe]

You might not find the SE’s as forgiving and I wouldn’t be surprised at all, if some SE’s take slobby coding as a sign of unprofessionallism and downgrade your ranking as a result.

Bull. Simply untrue and not bloody likely. Even txt files with no formatting at all get indexed. Search engines simply don't and never will care. To do otherwise would be assinine.

[g1smd]

When looking at Text files the spider does not have any

<table> tr> <td> < can you read this content?

errors to pick through.

[Natko]

Let me paste some exemples from my pages. And i'd like to test my quoteless pages on next Googlindexisation.
1. <a href=javascript:warp() class=w><img src=../img/some.jpg name=targetimage border=0></a>
2. <iframe name=iframe src=some.html width=476 height=304 scrolling=no frameborder=0 marginwidth=0>
3. <input type=submit value=some class=b onClick=parent.cal.location.href='some.php';>
4. <a href=# onMouseOver=changeimage(myimages[0],this.href) class=x>

All these works fine. I red w3c racom. It discusses SGML relationship to HTML, at the time it was written (?).
I focus this quteless topic in two dir: interpreter and Googlebot (the only one I care). If quoteless is no problem to a browsers, remains to test on Google.
But let me deducte some logic. Browser interpret an web content. This web content is also 'delivered' by Google too. I don't believe that browsers would interpret correctly something that in its basic would non be indexed becouse of that same 'errors' that browsers pass over. I appoligize for my bannana english, btw. That lead us to think that these two main 'tools' would use different standards for an very sensitive issue, I think. I don't believe in this.
All atributes on my pages are quoteless, except metas and some js. I repeat, I'd like to test them as they are now.

[R1chard]

Oooh!

<a href=javascript:warp()>
<img src=../img/some.jpg>
<a href=# onMouseOver=changeimage(myimages[0],this.href)>

I'd DEFINATELY put quotes in all of those. I don't care if it happens to slip through with one browser or SE. I'd always use them for anything non-alphabetical (ie #, /, %, brackets...)

If you ask me it even looks confusing and ugly too! And remember that the tags and attributes might even not display in the right colors in a source viewer...

[Natko]

I accept you suggustions, functionaly. Esteticaly, I like my bald tags :-)

[g1smd]

Any HTML attribute values with # or / or % or a space in them, and that are without "quotes" are non-valid HTML code.

In XHTML ALL attribute values MUST be quoted, without exception.

[vkaryl]

"Esthetics" have little place in coding. Compliance with standards is more important - while one does try to minimize code-bloat, there's only so much dreck one can delete without having compliance/validation go down the drain (maybe not in html 4.01 strict, but surely in xhtml!)

Since xhtml is the "next level" (well, okay - until something else comes along....), one should do one's best to code necessary html as close to that standard as possible. It's not particularly smart to defy the current logistical parameters just because one thinks it's more "esthetically pleasing"....

[ergophobe]

vkaryl,

Basically I agree and I try to follow standards to the greatest possible degree. My pages are almost always XHTML strict. However very often the highest performance is achieved through ignoring specs/standards and breaking rules. Reliability goes down, but performance goes up. That's true for tweaking a car engine, that's true for overclocking processors, that's true for mountaineering at high level.

I think the original poster is intent on maximizing performance provided the hit in reliability isn't too great. That's his choice. I wouldn't do it, but I don't think it's as black and white as you suggest.

The important thing is to be aware that you are ignoring a spec, why you're ignoring it and what the benefits and pitfalls are. Then it's your decision. In this case I think it's a bad decision because the payoff is so low. However, blindly following standards because they exist can also be bad.

Tom

[Natko]

Well, I can test baldness on SEs at all, but I can't test full cross-browser comp. If Google doesn't care (and we will know that for few days) and no one can find some example of browsers non-comp, that will be enought for me to leave it as they are now. My pages aren't xml, but html. I don't care if IE2 or 3 would missinterpret some color or something (e.g. w3c racom). How many times you saw "it works only in this and those browsers". And what to tall about xml. I don't know nowdays, but initaly, this tachnique, Google was't able to index, even it is w3c OK. I don't care for estetic preferences of someone else rather than me, about tags, too. The only thing I care is nowdays browsers and Google. My pages need not to have one bit more than neccesary

[ergophobe]

My pages need not to have one bit more than neccesary

I've got to ask. Any graphics on your pages? Flash? Font tags? HTML colors?

[Natko]

don't be sarcastic. I didn't say that my pages needn't to have anything. I have imgs, external css and js, dhtml, phpBB, phpFormMail, banner and stat codes, metas, text etc. etc.
Btw, this baldness (#, href and /) is OK with Lynx?
any non-sarcastic comment on this?

[ergophobe]

That's not sarcasm. That's an honest question. Nothing more, nothing less. In every post I've made in this thread I have tried to be polite, balanced, helpful and open-minded. I ask only the same in return.

The reason I ask is this. If you put an extra five minutes into optimizing a graphic, you can easily shave off an extra KB. Personally, I would leave my attribute values quoted and I would spend the extra five minutes on optimizing the graphics. If you're using phpBB, and some of the other things you mentioned, you have some obvious sources of major code bloat. A few extra quotation characters pale in comparison. It is perhaps the last thing I would focus on if I were trying to shave every possible byte.

Tom

[Natko]

OK. My imgs are optimised. phpBB is preprocesed and there is nothing I can do. I can accept that you would use quotes. I started this topic to see others what experience had in this materia. I have focused on G as a major SE, and latest browsers to see if quotes are mandatory or optional. I think this is the point, not personal preferences about estetics in tags. I didn't wanted this topic turn into discussion about my site and what does and what doesn't it contains. From this view I saw your post sarcastic.
For few days we will know the 'truth' about G.
We already know that with Lynx is OK.
Some validators recomand it becouse of w3c racoms. (do we need a selective racoms?)
IE5.5 doesn't care. And newests, I suppose.
By now I don't know nothing about others interpreters.
I'd realy like this discussion to conclude with something.
I also added some new pages, linked bald, to see if would be indexed. I will also compare my new listing positions.

[grahamstewart]

I have focused on G as a major SE, and latest browsers to see if quotes are mandatory or optional. I think this is the point, not personal preferences about estetics in tags.

It is not a matter of aesthetics.

The w3c standards tell us that quotes are MANDATORY on all attributes that contain non-alphanumeric data in HTML4.

Those are the standards.

Some validators recomand it becouse of w3c racoms. (do we need a selective racoms?)

Yup - thats because without them your code is not valid and a validators job is to tell you whether your code is valid or not.

[Natko]

Why we call them racoms? Becouse they are only racoms and if we follow them our pages would be 'universaly OK'. These are not mandatory rules, but standards.
If you prefer, I try to search how a non-standardised pages would be handled by latest browsers and by Google.
I feel like I'm the only currious webmaster moron?

[g1smd]

What is a racom anyway? ...(searched Google, none the wiser)

[Natko]

recom, sorry