Forum Moderators: open
[webmasterworld.com...]
Most of those reasons on the "bad" side are enough to rule out frames for the serious site.
and it is very very old-fashion (90's look) and unprofessional
www.google.com/search?hl=en&lr=&q=site%3Awww.webmasterworld.com+90s+frame&btnG=Search
"Ugly, 90's look" is nonsense. If you switch off scrollbars on minor frames, the initial appearance may be identical to a non-framed page.
I find scrolling of large windows to find key navigation points (which can be well disguised on some sites) extremely annoying.
No scrolling is necessary to navigate my site, I have no problems with ugly urls (they look normal) and no problem with search engines. It has to be said that achieving this is not especially easy, but it is possible.
Of course, frames would not be needed if IE supported {position:fixed}. It is possible to achieve this with sneaky CSS anyway, but I haven't tried this. The example code I've seen doesn't work properly on Opera (but looks like it could be made to work).
Kaled.
Several years back I had an opportunity to convert three different framed sites and see how the stats changed. One stat I watch closely is page-views per unique -- it's a good measure of stickiness. In these cases it was a three frame frameset, so just one hit from a unique visitor starts out with three pages. Nevertheless, in each case, presenting the same content on a "flat" page boosted the stickiness measure -- and sometimes nearly tripled it.
What I discovered in focus groups was that some people love frames and relate well to the concept -- it just makes sense to them. But a significant portion of the general public finds frames confusing and generally will not explore a framed site unless they have a very strong reason to.
My sense of this was that those who related well to frames tend to be good in abstract and hierarchical thinking -- two mindsets that web designers and tecchies usually have. But those who would never dream of learning to code also had big trouble understanding what frames are all about. It's just not the way their mind organizes the world.
As Jakob Nielsen said, "Frames break the metaphor of the web." If you've never read his classic article from 1996, check out Why Frames Suck [useit.com].
My site is software. My most popular program (56,000 downloads this year) has a download rate of nearly 80% per hit. I also have download rates of ~60% and ~30% for other programs. Judging by these figures, I don't think visitors have a problem with my use of frames.
I've read, in the past that you should never turn off scrollbars in frames. If you follow this rule you are almost guaranteed to end up with a mess - perhaps this is part of the problem - mindless people following a mindlessly stupid rule.
Kaled.
The reason why I switched back to non-frames in 2000 was because I had gained a better understanding of what a Uniform Resource Locator was. (ie. one resource, one web address).
The sites I converted were in the fields of medical software, spirituality, and crafts. The crafts site showed the greatest improvement, and the medical site showed the least, but still sales for them nearly doubled.
The previous framed versions of all three sites were well executed, but their stats were just limping along. The improvements we got just from flattening the sites, with nmo content change, were dramatic -- enough so for me to begin exploring WHY the changes were so great. So I now report on what I learned, but clearly the target market is a big factor.
I have no doubt that some markets are just fine for frames -- and I don't personally have any aversion to using frames and I also appreciate always having the main nav right at hand.
[edited by: tedster at 10:55 pm (utc) on Sep. 22, 2005]
Why don't I see many sites using frames?
Because frames are close to useless for public websites.....All the main reasons have been quoted above.
They are not so useless for private web applications....A several stats packages use frames as an interface; as do several specialised sites I'm logged on to right now.
So you don't see them, because:
a) even the public ones are hard to find because they index so badly in search engines; and
b) all the best ones are non-public applications.
I have seen many sites where javascript has been used (badly) to place a banner, etc in a fixed location. They're not using frames but I guess its ok.
If I showed you a snaphot of a page on my site, the only way you could tell it used frames is because the scrollbar on the right-hand side doesn't go all the way to the top. "My god" I hear you all say "how pukifyingly ugly that must be, a scrollbar that doesn't go all the way to the top!"
As for people not being able to understand frames, well, there are still a lot of apps out there that use old-fashioned .hlp help files (rather than .chm) and one very common feature is a fixed area at the top. "Ye gads, how can anyone actually use a help file designed like that - it would be pukifyingly ugly!"
Kaled.
in the hands of someone who knows what they are doing, these can smoke in Google.
My site is software. My most popular program (56,000 downloads this year) has a download rate of nearly 80% per hit. I also have download rates of ~60% and ~30% for other programs. Judging by these figures, I don't think visitors have a problem with my use of frames.
Since you (presumably) have no idea what the download rates could or would be if you DIDN'T have frames, I don't think these figures show much - although at first glance they look "good" (for some values of "good").
If that is the case I have some answers:
Because most frames were used for the wrong reasons by webmasters who did not understand what they were doing. (Been there, done that.) Many of those webmasters have realised their mistakes and corrected them. In some cases they have kept the frames and made them work better, but in most cases they have changed to something completely different, like CSS.
Because those people who teach people to build websites do not promote frames as much as they did five years ago.
Because there are more websavvy website visitors now than there were before. If a website does not function properly there is a much greater likelyhood now that someone will write and tell the webmaster and perhaps even suggest what is wrong.
Can someone explain to me why a user is likely to prefer it that way? Bear in mind, navigation of the site will be MORE difficult and the content and layout will be unchanged. For instance, having read the whole of the introduction to the product, they must first scroll to the top of the page before clicking a link to the download or license page, etc. How is that better for the user?
There is another problem with frames that I haven't mentioned - scaling but many website layouts break with or without frames. (Opera now resizes frames correctly!)
The main problem with frames is that they have been badly implemented. That said, html 4 still omits basic features that should have existed in html 1.0 (e.g. client-side include). The main problem with tables is that CSS specifications are inadequate. It still baffles me that there is no <menu> tag in html.
The objections people raise against frames remind me of those raised by C programmers against Pascal. They say pascal leads to bloated slow code and all sorts of other things. Well, that too is nonsense. However, it's always easy to spot Pascal code written by a C programmer - almost without exception it's slow and bloated. However, that results from ignorance rather than the failings of Pascal (of which there are many).
Kaled.
The main problem with frames is that they have been badly implemented.
Right -- but in the most basic deployment, using html only, the implementation problems are there inherently, requiring extra scripting to work around bookmarking and orphan page obstacles. This was not the original metaphor for html, and new authors inevitably have troubles.
Can someone explain to me why a user is likely to prefer it that way?
From interviewing many frame challenged people, I would say that the idea of one page containing several pages just doesn't connect at all. Whatever shows in their window is a page, and that's what their mental model of the web requires for them to function.
>a three frame frameset
in the hands of someone who knows what they are doing, these can smoke in Google.
Absolutely. Something about the Google algo really loves a page that has been unburdened of the basic nav links.
forcing other pages in the website into the frameset, using javscript?
Handy, Generic Javascript Coding [webmasterworld.com]
