<i> is not evil.

The truly semantic web developer might argue that the code contain neither [i] or [em], but:

Dave's office: [span class="book_title"]Web Form Design: Filling in the Blanks[/span] by [span class="author"]Luke Wroblewski[/span]

@dave: I would argue that is effectively creating custom tags that are both devoid of inherent meaning and ignore established practices that are known to and interpretable by assistive devices. No better that replacing [h1] with [div class="reallybig"] and [p] with [div class="contentblockstandard"]. (Which I've seen. I wanted to throttle whoever did that.)

It is more appropriate to extend an existing object with a relevent base meaning than to create new one objects out of whole cloth. The truly semantic web developer recognizes this. *grin*

It seems like you're arguing that em is inappropriate in this case because its meaning has been stretched, and that introduces ambiguity to em's meaning, and that lessens em's semantic coherence. So I don't understand how the solution is to "extend an existing object with a relevant base meaning" - the effect is the same :)

Also, I think the main reason you don't want to use [i] is that it specifies presentation rather than meaning. I think, too, that it's a bit too general to say "it's about distinction." All the typographical tags are about distinguishing some text from the rest.

Really? Are we arguing the title of the book? Sounds like you have a bigger problem.

Free Time > Work

@Nick If you're not spending at least 80% of your work day arguing about semantics, validation, operating systems, or indent style, you're not a web developer.

Not all fonts have italics but instead have obliques (most sans-serif fonts), so using the [i] shouldn't work in those cases, but it does for whatever reason.

Saying that a title of work should be 'italicized' is then the problem. It shouldn't be italicized at all.

@Dave, since they added "Senior" to my title I no longer argue, I "Delegate" arguments about semantics, validation, operating systems, or indent style.

Semantic purists usually say "you should use for that". And helpfully enough, the default appearance of that tag is... to italicize.

Look how modestly Egypt chimes in with what is otherwise a slap in the face to all other posters. Let me be a bit more abrasive..

http://www.w3.org/TR/html4/struct/text.html#edef-CITE sez "Contains a citation or a reference to other sources."

http://www.htmlhelp.com/reference/html40/phrase/cite.html sez "The CITE element is used to markup citations, such as titles of magazines or newspapers, ship names, references to other sources, and quotation attributions. Visual browsers typically render CITE as italic text, but authors can suggest a rendering using style sheets. Since CITE is a structural element, it carries meaning, making it preferable to font style elements such as I when marking up citations."

Robert (author) sez "The truly semantic web developer recognizes this. *grin*" @Robert: How cocky. You think you're correct. You shoot down another developer/designer while reinstating your elevated ego. Your example is wrong. Period. The truly semantic web developer does not recognize clearly..

Dave sez "@Nick If you're not spending at least 80% of your work day arguing about semantics, validation, operating systems, or indent style, you're not a web developer." @Dave: When you're sufficient in semantics, validation, & operating systems you understand them enough to avoid silly arguments entirely on the basis that there is usually one correct way to interpret these concepts properly. When a fool surrounds himself with fools he's sure to discuss foolishly.

Gary sez "Not all fonts have italics but instead have obliques (most sans-serif fonts), so using the [i] shouldn't work in those cases, but it does for whatever reason.

Saying that a title of work should be 'italicized' is then the problem. It shouldn't be italicized at all." @Gary: It should be italicized. Here check it out: Robert (author) sez "Thanks to centuries of typographic experimentation, evolution and standardization, we already have a method for accomplishing this: Italics are used to set out book titles as distinct from the rest of the text." I do agree with this statement. Don't let a font designers ineptitude change your worldview of typographical design.

Nick sez <<@Dave, since they added "Senior" to my title I no longer argue, I "Delegate" arguments about semantics, validation, operating systems, or indent style.>> @Nick: Since "they" (hah) added "Senior" to your name and gave you a bump in salary you've now reached the heightened status of "delegating" the same bullshit arguments you never should have participated in in the first place.

@Egypt: Spot on brother sister :). Love the design.

Why the harshness? Standardistas that think they are but aren't are damaging to the overall ecosystem of standards -- especially when you blog incorrectly and troll the comments section incorrectly.

This is a good point and I agree. I hope those with an interest in the semantic web take a look at the MLA writing standards and other such bodies of work. I think the semantic web is a great idea, but we should perhaps standardize the categories like we standardize MIME information. (this is perhaps already in progress)

Also, @Nick, since when was "free time > work" a bad thing? :)

I concur with Egypt above... isn't the most semantically applicable tag for titles of printed media?

gah, lame CMS.

*ahem*

I concur with @Egypt above... isn't the cite element the most semantically applicable tag for titles of printed media?

I would disagree -- the distinction between i and em is imaginary. No meaningful distinction can be made. If you can't see the difference then no one is checking the difference, and if no one is checking the difference then you can't trust the difference. If you can't trust the difference, there's nothing functional you can do to distinguish (including changing styles to distinguish between them). If there's no functional differences then the distinction is pointless.

http://blog.ianbicking.org/2007/08/14/of-microformats-and-the-semantic-web/

Marc, Ian, you are wrong.

i says, do italics. em says emphasis.

One is style, one is semantic. One says, make text italic, but why, the other says, emphasis something. (Doesn't say why though!).

The guy saying that you should create a class to distinguish is utterly correct.

[class="title"]

You can distinguish it on screen in one way, on print another, and for screen readers, you can add emphasis if you want, or not.

@Mike: "gah, lame CMS."

By that, I assume you mean it rendered your HTML. Crazy, I know! You added an HTML tag, the lame CMS outputs exactly what you entered, and your web browser renders it.

I fixed your comment, replacing your raw html with entities.

@Jack

class="title" would be correct, as long as it exists as an attribute of the cite element.

Jack: to what end? When do you choose to make the distinction? Should I markup my nouns as nouns? Should I distinguish between different kinds of emphasis? Should I make note of why italics are used in other cases, like a foreign language? To what end?

Semantic markup is interesting only if there's something you can *do* with that information. I assert there is nothing useful you can do with the distinction between i and em, except perhaps for some exceptional places where you have a walled garden of content and you've created and maintained a distinction. In the rest of the world there is no reliable distinction between the two. You cannot use the distinction to any end because the point of markup is to determine the author's intent, and there's no reliable intent that is communicated through this choice of tags. Sure, you can come up with rules about when to use the different tags, but you can't come up with reliable rules about how to interpret the different tags. You can't just make up rules and have people automatically obey them. Without a plan to get people to write to your rules, your rules are meaningless.

Brutal. Just brutal.

@Patrick: How condescending. No response? @Robert: No response?

Brutal indeed. Deserving? Yes. Why?

Let me recap that: The tag is semantically inappropriate in this context.

Correct.

Use instead? Incorrect.

No mention of CITE.

You've unintentionally infected Reddit, Google, and likely the blogsphere as a whole (via Technorati, etc..) with misinformation and have not posted a correction regardless of comments found below. That's just irresponsibility.

On that note, I agree that designers should continue to use for the reasons not related to citation or emphasis contained here: http://en.wikipedia.org/wiki/Italics#When_to_use

As for the arrogance previously noted in regards to Robert and Dave.. It seems you guys work together.. If this had been a personal blog I would have ignored it entirely. You guys represent a web design company and your comments and that little iMarc logo command a certain expectation of quality.

And now for Ian's take on this damned little tag: what's right is what's known and understood. We can all sit and pontificate the true meaning but what it comes down to is how Google, Firefox, et al. interpret them..........

**yawn** Correct your post or stop posting. That's my opinion. Brutal? Sure.

As for the CMS, am I supposed to encapsulate my paragraphs in P tags to get them to break properly? The comment input field should parse and render appropriately. I can't believe I have to use manual encoding..

Dan SimpleBits wrote a similar article in 2003: http://www.simplebits.com/notebook/2003/09/25/simplequiz_titles_conclusion_sort_of.html

As he mentions, "if you’re simply stating the title of a book, are you really citing it?" I'd tend to say no.

Another interesting tidbit from the simplebits comments was someone saying that he was from Estonia where the standard is to put book titles in quotes, not italics. In my mind, this gives more credence to using a tag that describes what the content is — — The presentation could theoretically be localized.

@Angelo You used the tag in your response. That was messing up the display. The CMS didn't properly identify that tag as an inline element and assumed you were using block-level tags. If the CMS notices block-level HTML, it won't add breaks for obvious reasons. I fixed the backend.

My goodness, the tag.

Angelo, you are the Leonardo da Vinci/Tucker Carlson of blog commenting.

Hat's off.

@Angelo: Despite your abrasive ad hominem attacks, I've enjoyed your contributions. As for not correcting the post ... I want to let the stream of (unexpected) comments settle down before deciding what to incorporate. I do plan to post an update; whether someone posts it to reddit is beyond my control (I was quite surprised to see it picked up).

And as Dave notes, cite is not really the appropriate tag to use for a book title. It is commonly used for this, but I do not feel that is a proper use of the tag's original design intention.

However, I must note that the W3 XHTML 2.0 workgroup feels otherwise, and explicitly use cite for a book title in their cite example, in conjunction with quote for the excerpt cited.

And, the HTML5 spec is even more specific, noting that cite must be used only for the titles of works.

However, this is getting bogged down in a single example -- which, obviously, was poorly chosen, because I was wrong. :-) Other accepted uses for italic which are not accounted for in the various HTML specs include the names of ships, foreign words, use of a word as an example of itself rather than for its semantic content, introducing new terms for definition, and Symbols for physical quantities and other mathematical variables (Italic type - Wikipedia).

i remains not evil.

A final note (well, for the moment). The q tag is deprecated. Arguably you broke the CMS, not the other way around. ;-)

@Dave:

As he mentions, if you’re simply stating the title of a book, are you really citing it? I'd tend to say no.

I suppose I could go either way on that one. One could argue that you are, in fact, citing the book title even when stating it without following quotations. cite also has the benefit of being rendered in italics in most (?) browsers — by default and without additional CSS rules. How many I could not tell you definitively, and that would be an interesting tidbit of information.

Your opinion is to tend towards "No". The author of that SimpleBits article is a bit more flexible in his judgment. He seems to be swayed by the fact that "most" browsers italicize contents. His skepticism seems to lie in the uncertainty of widespread browser support. It's been 5 years since that post and it is my opinion that browsers support the

I downloaded the Estonian language pack to my FF 2 and much to my surprise it did not render the citation properly (using quotes instead of italics). If I were Estonian and it bugged me I'd file a feature request using a Greasemonkey script in the meantime. To suggest that deviating from the semantic richness of in place of custom CSS brings us back to Ian's post. Neither Google nor Firefox will know what it means. To further this concept. Say I was a researcher. I'd surely tune my Firefox to visually acknowledge citations (via double underline, icon to the right, icon in the status bar, etc..) allowing me to quickly run searches at Amazon.com, OpenLibrary.org, and research databases I'm a member of. This is clearly possible by use of the tag. This is clearly impossible by use of custom CSS *unless* we standardized a set of classes for this purpose (like Microformats). But wait.. it's already been standardized! DRY.

@Robert:

And as Dave notes, cite is not really the appropriate tag to use for a book title. It is commonly used for this, but I do not feel that is a proper use of the tag's original design intention.

Again I'd err on the side of using since it's machine readable. We're not talking about English 101 and parenthetical citations here: http://www.barnard.edu/english/reinventingliteraryhistory/plagiarism/citation.html

However, I must note that the W3 XHTML 2.0 workgroup feels otherwise, and explicitly use cite for a book title in their cite example, in conjunction with quote for the excerpt cited.

And, the HTML5 spec is even more specific, noting that cite must be used only for the titles of works.

What we can all take out of this is a) the difficulty in coming up with standardized solutions and b) the lack of coherence in the standardization community. I like how you referenced XHTML 2.0 and HTML 5, both being on the forefront of markup standards. But wait.. which one do we use? Keep in mind that your post is rendered in XHTML 1.0.

@Patrick:

Speaking of ad hominem attacks.. It's so clear you're "that guy". Go watch a South Park, ice that sweet bruise, and leave the discussion to the adults.

After work you'll find me: Watching the Sox, shooting things on Xbox 360, hanging at the Fish.

@iMarc:

Dave sez..

@Angelo You used the tag in your response. That was messing up the display. The CMS didn't properly identify that tag as an inline element and assumed you were using block-level tags. If the CMS notices block-level HTML, it won't add breaks for obvious reasons. I fixed the backend.

I see you added: Accepts and renders HTML. If you include any HTML other than inline elements, you’ll also need to include your own paragraph breaks. Prior to my complaint you didn't even specify that you accepted HTML nor that it was necessary (I still don't even know if it is..).

Robert sez.. A final note (well, for the moment). The q tag is deprecated.

Damn. I was so close to apologizing for my abrasiveness.. The tag is not deprecated for XHTML 1.0, again, in which your site is rendered.. Care for a more formal spec? http://www.w3.org/TR/xhtml11/doctype.html says ... the XHTML 1.1 document type ... does not contain any of the deprecated functionality of XHTML 1.0 nor of HTML 4. You must be confused with XHTML 2.0's tag. Neither we (the community) nor you (iMarc) are there yet.

Word of advice: scratch HTML input in the comments entirely and use markdown instead. I love HTML. I hated having to tag and encode my comments manually... Furthermore, I don't even have a "Preview" facility. Do I proofread this in another tab before posting? Well here goes; hope it all renders properly.

Never would have imagined would cause so much grief. :)

@Angelo: Never would have imagined would cause so much grief. :)

No kidding :-)

Regarding Markdown, you might want to check out the Wysiwym Markdown Editor project. (The author is even working on a version to be released under the MIT editor.) It's a possible way to bridge the gap between "what i type" and "how i want it to look"... (Markdown is great, but not many people know how to use it yet.)

My own team page, used against me. This is why we should only list v-cards.

@patrick with mobile phone numbers, IM handles and home addresses, 'natch.

All this grief from a stand-alone web page owner with a self-referencing link.

@Robert: Neat. It looks like Markdown + Showdown with keypress event handlers. I'm in the need for just this at the moment and was about to mash the two up in a jQuery plugin to achieve the same affect. Thanks for the input.

@Nick: As director of development I would have expected you to be a bit more wise. The page contains the perfect amount of information to identify exactly who I am and how to contact me. It contains my name, affiliation, general location, and two preferred methods of contact. It also contains my hCard and serves as my OpenID delegate. I don't feel the need to say half the crap you guys do. That's my opinion and you're entitled to yours. However, in honor of your comment and on behalf of this fabulous thread I've gone and updated my webpage. I suggest you all take a look -- Patrick in particular. :)

I find it comical that several of you are fathers. From your words and your pictures you guys seemed to be closer to my age..

Now in an attempt to finalize things here please see the Reddit comments for your post, this thread in particular.

Go ahead and click "Reply" to either post and click "Help" beneath the resulting textarea. Now scroll up through the entire comments section of this post and give me one example where those 5 features would leaving you wanting more control over markup.

Now "Cancel" the reply and read the comments. The reason why I made my first post was that your words, Robert, may have been specific to an audience that you are comfortable with but were, in actuality, cast upon the world as a whole. Some kid subscribed to your feed here in the states (who won't see comments) or even some kid in China who translated your page could have been mislead by your post. Furthermore, 1,000 people (keeping with proportions of the 80/20 rule) could stumble upon this post (no pun intended).

I'll leave you guys with a quote from Tim Berners-Lee himself. Please tell me you've heard of him.

By recognizing the Web in such a significant way, it also makes clear the responsibility its creators and users share," he continued. "Information technology changes the world, and as a result, its practitioners cannot be disconnected from its technical and societal impacts. Rather, we share a responsibility to make this work for the common good, and to take into account the diverse populations it serves.

Finally, something even I can appreciate.

As a director, I'm adding you to my list of future hires to pass on. Also, IE hates your new site, lovely mime types.

Word. I hate IE. See how that works?

I mean did you even read the above comments? Do you really think I'm that incompetent? Really.. Guess what I did when I was modifying the source: changed the file name from index.html to, wait for it, index.xhtml. Why, you might ask, would I have made such a significant change to simply add an image and a list?

Here's the best part: I wanted to see how many of you use IE as your default browser. You know, "eww, it's even uglier than it already was" or "I can't believe you call yourself a designer, Mr. Salvador Dali." So, the way I see it, you either a) use IE as your default browser *ahem* or b) tested my site using multiple browsers to try to catch me on something, anything. Well kudos to you Mr. Director, I caught you. In case you're still not convinced, let's glance at how I'd fix this in a production-level environment whereby I couldn't simply revert the file's extension:

if 'MSIE' in web.ctx.env['HTTP_USER_AGENT']:
    web.header('Content-Type', 'text/html')

You wanna play games? Your page doesn't validate XHTML 1.0 Strict. Good thing you just straight serve text/html and not even worry about it.... Pathetic.

Enough? Of course not since Nick sez..

As a director, I'm adding you to my list of future hires to pass on.

I just lost what little respect remained for iMarc as a Design/Development firm. You might take a page out of Robert's book and learn that a little bit of humility can go a long way. That was, after all, the motive for my initial rant. Should I even have to explain this to you?

So now, unless I'm propositioned with another dose of tomfoolery, I'll leave you and your "Senior" title to delegate arguments ... about indent style amongst your team. Do the world a favor and don't blog about it. In other words, I'd never in a million years approach you or your company for a job.

You couldn't have left it at fun and games and serious discussion of semantics? Now you're going to have to kill this post and cover up your tracks. Guess I'll just have to cache it for memory's sake. I'll leave the URI in tact as it'll be a painful reminder that indeed can be very, very evil. :)

@Angelo: two items. First, your research on who does and does not work at iMarc could be a bit better. Second, while oblique ad hominem attacks are amusing at first, you're coming dangerously close to the line of being abusing. Please be considerate of other commenters. You've made some very insightful comments; please don't ruin it by getting personal.

So you're telling me that Nick has followed this thread from inception and just impersonated iMarc's Nick? How bizarre.. OpenID where art thou. Final post. I'm out.

@Angelo: Surely you're not the only Angelo in the world... Name collisions are common.

Yup.. At first I thought all commentors were random people from the web. Then I realized several were from iMarc. I considered Nick's Really? Are we arguing the title of the book? Sounds like you have a bigger problem. Free Time > Work to be a jovial inter-team comment from "Senior" == "Director" to staff.

Boy was wrong. You were right, Robert, had I discovered that earlier much could have been avoided..

The sincerest apologies to iMarc and especially you Nick Grant. You too Patrick. :)

Received a nice email this morning, respectfully replied and ended. This is the web, people get heated and make paragraph attacks on anything not directly inline with their thinking. Stooping to attack browsers, speech, word choice and other things is how the web lives and breathes. We are faceless entities free to make any and, in most cases, all bold statements without fear of retribution. Being "attacked" in a blog post lives on my scale of actual hurt about the same as being bombarded with marshmallow's in a nerf factory. We're free to say and do whatever and I take full responsibility for all of my statements. Funny story about my site though, being coldfusion, it isn't exactly text/html. And yeah, I don't render XHTML strict. My choice.

For the sake of transparency and since every single one of your posts thus far have been sarcastic, here's what that nice email contained:

I suppose I brought it upon myself. You didn't have to run with it!

I hate to shoot a dead horse on someone else's farm but let's recap your intellectual input throughout this discussion.

Really? Are we arguing the title of the book? Sounds like you have a bigger problem. Free Time > Work

@Dave, since they added "Senior" to my title I no longer argue, I "Delegate" arguments about semantics, validation, operating systems, or indent style.

All this grief from a stand-alone web page owner with a self-referencing link.

As a director, I'm adding you to my list of future hires to pass on. Also, IE hates your new site, lovely mime types.

You offered absolutely zero value. One might even go so far as to say your presence had a negative impact on the discussion.

Harshness aside I know that I at least attempted to cast some light on the discussion. The last couple of posts were harsh for all the wrong reasons, you see, for I made the horrible mistake of grouping your worthless input with that of the iMarc team.

It turns out iMarc's team had very good things to say not to mention I agree with the contents of the original post and have myself learned something from investigating things to the extent we did.

Your e-mail response..

Consider this a free professional lesson. You can say and do whatever you want on the internet, because it is faceless. Until you put a face to it. This is the real world, people remember things like being slighted, brought into question and in general treated poorly or unfairly.

As my work on the web revolves around digital identity this thread has been more than just a professional lesson -- it has reinforced Who Is the Dick[/Nick] on My Site? ten times over. Excellent powerpoint presentation on Identity 2.0 for anyone interested.

As for the faceless part.. I posted my face. My demeanor didn't change. I was actually expressing humility which I have good reason to believe was received well. You, on the other hand, could have been forthright and provided your full name saving me the research (DNS, MySpace, ...).

Your marshmallow/nerf analogy speaks volumes on your behalf. So do the words you so arrogantly stand behind. As for your site.. Coldfusion's your backend. You're serving vanilla text/html.. with tables and form fieldsets strictly for presentation. At the risk of sounding whiny whyyyyyyyyyy are you involving yourself in a discussion of web semantics?

I felt the need to defend that last statement and to one last time reiterate that the iMarc team has shown personality, compassion, and a rigorous respect for web standards and design. I apologize one last time for the conclusions I jumped to.

@Nick: If you really feel the need to take this any further a) please don't and b) keep it in the mailbox.

i is evil. For distinction correctly you'd use span/div. Using i would be the same as using font tag.

p.s. don't use underscore in naming classes/ids.

[i] is not evil, [i] is a tag. One could say that its semantics are a bit strong for use and in that case I concur with Jack, above.

You guys are nuts. Its about the user, not the technology. [i] and [em] should both be acceptable because there is no valid reason for them not to be. Standard nazis. Just too funny. The big problem with defining only one right way to do things is that there is no why. Yesterday's standards are today's violations as today' will be tommorrow's.

I've been building since connections were 2400 baud and it always changes. Yes, my code passes validation. No it doesn't have browser detection cause its not necessary. [i] is not evil [em] is not evil. its all just a way for you guys to say the other guys suck.

@watcher

I think there is some sort of misunderstanding here. The reason standards and semantics are talked about so much is mostly the end user, and partially those who must maintain the site. I think you'll notice that most of the web community pushed these topics because they believe it makes a better experience for everyone.

Sites built with standards are good for end users because no matter what the web browser, it should be able to display the content is some sane way. Sites built with standards are good for maintainers because they are more likely to not break when new versions of browsers come out.

Sites built with semantics are good for end users because intelligent spiders and other computer programs can actually extract the meaning of content by the tags. Sites built with semantics are good for maintainers because things tend to be more clear. When you don't use presentational adjectives in your CSS class names, is it much less likely that changing the style of the class will confuse future maintainers. Nothing is more confusing than a .wide_column that is narrow or a .blue_box that is red.

@Robert

I would tend to disagree with you that the tag should ever be used in HTML. It is never semantic, because it only conveys style. If you need something that is not going to be emphasized, the proper semantic way to mark it up to to explain what it is. I would mark up the book title as . This way you can actually style all book titles with whatever works on your site.

Perhaps you will have a different background color or increase the font size in addition to making it italic. These would be impossible with a simple tag (i.e. no CSS class). If you then add a CSS class, it is entirely possible someone would change the font-style of i.book_title at some point to no longer present it as italic and then you are confusing maintainers.

@Will Bond

I totally agree with what you have said. Like I said, My stuff is valid and works on all browsers without detection or CSS hack of any sort. I do use [i] tags because flash doesn't know what [em] is. Nor does it know what [strong] is and we use some flash elements. It was not my intention to imply that proper proper class naming or any other good coding practice was bad. My point is two fold. 1. Enough whining. Enough with the hard brick wall. 2000 rows of data pouring into a site and some clown's complaining that it doesn't pass validation because one page has an unclosed tag. In the past 15 years, "the way it MUST be done" has changed about 15 times.

and 2. Stop Browser whining. Firefox is good. IE is good, Safari is, well, Safari and good. I don't care what anyone uses.

Just write good code that works in all browsers with no hacks. I have 80 domains out there not one single browser specific CSS hack anywhere. Maybe its 'cause I'm old but seriously 35+ comments on a italic tag. jeez.