<DIV> and browser compatability

In addition to mariush points,

<DIV> makes -

• Loads pages fast
• Lower hosting cost
• make redesign more efficient & less expensive
• Maintain visual consistency
• Better search engine results
• Make site more accessible to all viewers and user agents
• Give a competitive edge

In fact, It saperates content from presentation.
-

 

You're more or less right with most of your points Dakait, but I'd be careful with blanket statements like those. I'm not sure why you think hosting costs would be lower if you used divs.... why is that?

btw you can use divs with tables, it's not one or the other.

 

I can explain why, though the difference has everything to do with the amount of traffic you experience.

We did a little proof of concept regarding a company Intranet. The average page contained 27 kb. of data, and it was almost all HTML-based style markup, very little CSS. We reduced the average page size down to ~4 kb., and then we did the math.

A company of ~14,000 employees use the Intranet on a regular basis. X# of those employees are accessing it from remote locations, using expensive airbus minutes. There are international employees and distributors who are relegated to 28.8k modem connections still, and sometimes worse, depending on where they are located. When all of the analysis was done, the savings came out to ~$1.2 million per year, based solely on four hours of work converting their Intranet templates over to tableless layout (with tables used for tabular data still). Of course, there's a huge time investment in converting all 60,000 pages of the Intranet over to a new template, and it's not unlike corporate America to drag its feet on stuff like this. But bandwidth was a significant portion of this. We talked with the company who provides the pipes and backup connection, and bandwidth savings would account for tens of thousands of dollars of those savings on a monthly basis.

I'm not at liberty to discuss details, but I just wanted to make the point that using proper markup is not only beneficial to the end user and search engines, it can have a significant impact on the business itself too.

 

And of-course, CSS style sheets are cached by browsers, so images download only once if contained within the style sheet declarations (so long as those images are used repeatedly throughout the web site). Page load speeds of CSS styled web sites are predominantly much quicker than straight html sites (compare them in Alexa for instance, who will show significantly faster speeds for CSS based layouts than html with tables, etc), not least because there is less code and more caching.

Hosting costs would of-course be lower, how much lower depends on the amount of traffic, and as the_pm states - large amounts of traffic can make significant savings.

 

Wow the_pm, looks like you did your homework! That's quite interesting.

 

Thanks

Yes, we factored in everything, including the amount of load time time saved v. the average hourly wage (and estimated hourly wage of salaried employees) to determine the opportunity costs as well. This part of the analysis went something like this (I'm making up the numbers to serve the example - again, I can't release real numbers):

Avg. # of pages viewed per employee per work day with Intranet access: 15 (includes employees who do not access the Intranet at all, which skews down the number dramatically).
Total number of employees with access: 10,000

Avg. access speed: 40 kps (airbus/int. employees skew this down dramatically).

Time spent loading pages in a given day: 81 seconds
Avg. hourly wage: $20
Payroll cost: $0.45
Payroll cost for time spent downloading ($0.45) * total employees w/access (10,000) = $4,500/day.
Total work days per year: ~250 (minus weekends and avg. 2 weeks vacation)
Total work days * daily payroll costs = $1,125,000 in time spent loading pages per year by all employees.

After code reengineering
Time spent loading pages after recoding: 12 seconds
Avg. hourly wage: $20
Payroll cost: $0.066
Payroll cost for time spent downloading ($0.066) * total employees w/access (10,000) = $660/day.
Total work days per year: ~250 (minus weekends and avg. 2 weeks vacation)
Total work days * daily payroll costs = $165,000 in time spent loading pages per year by all employees.

Cost savings: $960,000/yr.

There's a lot more to the analysis than this, but I just wanted to give some highlights. There are holes in it, other considerations that are taken into account with the full-blown analysis, but I just wanted to give you the flavor of the whole thing.

That's just one area in which there was cost savings. Less bandwidth means savings. Less hard drive space/less servers means savings. Less IT staff to manage less hardware is savings. Etc. etc. It's amazing how a little code reengineering can have such a big impact on a business.

Again, those numbers are baked, but not at all unreasonable for a medium-large corporation.