None of the things listed in this poll will cause your web pages to break. In fact, only a couple are even mentioned as being a bad idea by the W3C. But all of them can cause some web designers to cringe at the thought. In fact, things like quoting attributes or not writing HTML in lowercase are things that up until a short time ago, I felt were very important, but now I do all the time. While others, like using tables for layout or using images for text instead of plain text, I still feel are bad. What pet peeves do you have about HTML?
I wouldn’t say anything annoys me as such, but some things I would prefer not to use eg frames.
I think the business of validation is a tricky one. I had a client who wanted code validating because he thought it would look more “professional” even though I though the site stopped being creative.
It’s something to always consider but I don’t think purism brings about creativity as regards web design, as all things that can be seen as thinking outside the box may not be part of HTML validation.
Well, I’m not sure I agree that writing valid code stifles creativity, but I know that some people feel that way.
But I can say this: if you want to get creative, then by all means do so. But don’t get frustrated when there’s a problem and when you ask for help you’re told that your HTML doesn’t validate.
In other words, when you want to get creative, you need to take the consequences of that creativity and figure out creative solutions to any problems on your own.
Ha Ha you needn’t of replied Jennifer, I knew what you would say
“… validating … even though I thought the site stopped being creative”
Sorry “charlie”, but standards and validating against them makes code simpler to write and maintain.
I have heard that “creativity” whine against standards for years. It doesn’t wash. Once you get comfortable with the standards you will find reading and writing the code to standard much easier. You will find it much easier to read code that someone else has written if it is done using the same standards you are used to. You don’t have to focus on the mechanics of deciphering and writing the code, freeing you up to concentrate on the logic and design of the code.
If the site is not creative, it is the fault of the lame content the developer put into it, not the underlying code used to display it. Don’t waste your “creativity” on the code which is not seen, focus on the visible content! “Creative code” often turns out to be ‘bad’ in the long run. It is hard to read, hard to maintain, relies on obscure tricks (that 1 in 1000 people are familiar with) to work, so it does not adhere to standards. And it will fail in future browsers that enforce standards more strictly.
Hi rohn, I certainly don’t find validating code difficult, it’s very easy, it’s sometimes just nice to bend the rules like in any pursuit.
Of course it’s best to stick to validation as much as possible but for eg if you start using flash you can kiss goodbye to validation but would you stop using a technology because it doesn’t quite validate yet is best for a site?
I’m still learning html so that is not too familiar to me yet
Invalid HTML is the professional equivalent of bad English. You may think to yourself, “I read that HTML book and now I are a gud Web devilopper”, but any clients you may briefly have would not share your enthusiasm. (Or if they did, they are not clients I would care to have.)
Other bits, like frames and big blobs where text could go in a modern page, simply scream that you hearken back to the Bad Old Days of the Web, circa 1996, when there really wasn’t much choice but to use such primitive kludges. Fortunately for the rest of us, that was 15 years ago — time well spent moving the goalposts so far forward, they might as well be on a different continent now. If you choose not to move, fine; your would-be clients likely will.
Using tables for content is an egregious case of the mentality alluded to in the “frames and big blobs” paragraph. Tables are for tabular data, full stop. Unless your page is an Excelish mass of gridlocked data, with nothing else so far as the eye or the browser can see, you’re just making life harder for everybody. You’ll work harder maintaining and reusing your table-oriented magnum oopus. Your clients’ bandwidth costs will be higher, and their site performance and search-engine rankings lower. And you still will blow your schedule to little bloody bits making things look exactly right in IE.
Our Craft, like any craft, is only as strong as its weakest practitioner. When I first got into the Web, fifteen years ago, I figured that we were 15-20 years from something approaching a reasonable profession, in the true sense of that word. I’ve since revised that estimate to 20-50 years — from now. Things like this are a big part of why.
I checked “Some other (annoying) technique”, and since a comment was asked for in this case, here it comes:
Using a web creation tool that clutters the body text with formatting every, or every other, sentence. I have seen page sources where the body text seems to contain 15% text (to be displayed) and 85% formatting codes, e.g. font sizes and placements. However, I think Dreamweaver is OK.
Much of this type of Malformed CSS and HTML as well as pure crud is due to building Web Sites using Microsoft Office. The crud (invalid propratery CSS and HTML CSS tags) is due to “Round-Trip” tags that allows the file to be read back into Office as if it were a .DOC/.DOCX file (ie: The Crud Tags are documentation of the Word Formatting). When you save the file as HTML there is an option to not generate these tags thus saving space and helping the Browser to scan/interpret the file.
As to the Malformed CSS, while the CSS is totally valid, there is too much of it. Instead of declaring a P definition with the default settings (and thus making the P tags just , the default is declared as a class and then added to EACH p tag as . There should only be overrides using or use of class= with no style or class for the default style. Also even the 2010/2011 versions do not emit a Doctype declaration (which in 2011 is BAD/Invalid/Malformed HTML in-and-of itself).
For myself, I classify validation errors as either fatal or non-fatal.
A while back one of my clients was upset because his competition was pushing a new website locally. When I looked at the validation there was a lot of errors. I don’t remember what they were but the site looked fine in IE, a little off in Safari and Opera, but in FireFox, all the content was missing, and the top and side navigation was a mess. At that time FireFox made up about 20% of the viewer audience. So that site lost about 1 in 5 viewers. These were fatal errors.
Forgetting to close an image, meta, or input tag (HTML) will either give an error or at least a warning, but I call these non-fatal errors.
Oh, for certain many 3rd party links like Google Maps, or Mapquest will cause errors on a page, but first, there’s almost nothing you can do and second, they are usually not fatal.
(Many times if you find an un-escaped ‘&’ and change it to & it will fix the error.)
Validate and check your pages in as many browsers as you can. The results are worth the effort.
I’m not very experienced with web page creation–just starting to read some tutorials. But I can say that any layout that forces me to scroll from left to right and back to read is a sure guarantee that I will not stay on the page for very long. What’s wrong with word wrapping?