Page 2 of 3

Posted: Fri Apr 02, 2004 4:38 am
by malcolmboston
1kb to 3kb of text links!?

its not worth the hassle to be honest

Posted: Fri Apr 02, 2004 4:58 am
by twigletmac
malcolmboston wrote:1kb to 3kb of text links!?

its not worth the hassle to be honest
What about the additional load on the database server?

Mac

Posted: Fri Apr 02, 2004 5:27 am
by JayBird
i agree with Twig on this.

You can't just discount frames from the off!

You need to come up with the best solution to suit your needs and i think the developers of phpMyAdmin were totally correct to use frames.

Mark

Posted: Fri Apr 02, 2004 5:54 am
by redmonkey
Roja wrote: First and foremost, they break fundamental modes of web-browsing. Back/forward buttons don't work consistently with normal web-browsing paradigms.
Having poorly implemented or non existant navigation can lead to problems with users using the browsers back button when viewing a framed website. It is not so much a fault with frames but more so poor design. Relying on the browsers own back button is bad practice/design IMO
Roja wrote: It makes printing difficult - the default prints the current main frame, not the full frameset (if that even, different browsers implement this slightly different depending on design).
All browsers I have come across (that support frames) give you options on which frames you wish to print, if the user does not know how to use their own browsers print options then they deserve to get unexpected results.
Roja wrote: You often can't bookmark any page other than the root - again, depending on browser implementation.
This can be quite useful, having your users entering your site at one single point could be useful especially if that entry point contains your latest news , updates or any important annoucements.
Roja wrote: Search engines can't index them correctly - leading to poor ratings.
What about the noframes tag?
Roja wrote: Depending on the design, users might get caught in your frameset, even when clicking on outside sites.
Again, that is down to poor design not the actual use of frames.
Roja wrote: If the browser doesnt support frames, it will literally replace it with nothing - often completely breaking the design. ie, it doesnt 'degrade' gracefully.
What about the noframes tag?
Roja wrote: So, there are lots of solid reasons not to use them, and few good reasons to do so.
Personally I don't feel that your arguments against frame usage are solid.

Posted: Fri Apr 02, 2004 6:54 am
by MadButch
I'm not using output buffering yet, but I did get the loading functionality working.
What I do is:

- Before the loop starts that does that heavy work, I display an animated loading gif.
- After the loop I have a little java script that hides the image.

As you can see, it works great :)
(imo off course... the frames part comes next)



I had no clue that frames where so not-done. I didn't use them because I'm a lazy coder, but I am a "nooby" so you can throw me in that category.

I'll look into style designing with css. I tried to look a little into it, but I didn't yet understand how it works (haven't found a good tutorial on explaining it yet).

Re: My first PHP site, pointers/tips/tricks/feedback welcome

Posted: Fri Apr 02, 2004 7:43 am
by MadButch
Zay wrote:
MadButch wrote: I also would have liked it if the PHP files didn't show in the status bar of the browser, but never found out how to do that (didn't look that hard either :oops: ).
you mean with a link?
No... When the page is loading.
I set the status to "Loading, please wait..." but if a user moves his mouse over one of the visible links, and then off it, the long line containing the full adress is shown again in the status bar. (99% sure that it is because the onMouseOut event sets the status back to an empty string)

If the status bar had a history, you could ask it to show the previous message :) (which is probably doable if you use OO design, but I'm not that far yet).

Posted: Fri Apr 02, 2004 8:54 am
by Zay
if you use the output buffering (I do not know the excact fuction names atm, haven't worked much with it) there won't be any links on output untill you flush the buffer...so just set

window.status = 'loading...please wait';

before you start the buffer... then the last line before flushing should be

Code: Select all

<?php
echo "Windows.status = ''";
?>
this is so it doesn't show loading while done loading.
I think that should work. if it still isn't working I'll see if I can setup some testcode for you next sunday (won't have any spare time untill then)
good luck!

Posted: Fri Apr 02, 2004 8:54 am
by Roja
redmonkey wrote:
Roja wrote: First and foremost, they break fundamental modes of web-browsing. Back/forward buttons don't work consistently with normal web-browsing paradigms.
Having poorly implemented or non existant navigation can lead to problems with users using the browsers back button when viewing a framed website. It is not so much a fault with frames but more so poor design. Relying on the browsers own back button is bad practice/design IMO
So - your opinion is "While you are right, users shouldnt expect the browser to do what they want in my opinion". While the site CAN work around it by having stronger/better navigation, it doesnt change the fact that it is taking away an established paradigm THAT USERS USE OFTEN.

You are entitled to that opinion. But when Aunt Tillie wants to visit your-amazon.com, and cant get around because the back button doesnt work? Your loss, my gain.

redmonkey wrote:
Roja wrote: It makes printing difficult - the default prints the current main frame, not the full frameset (if that even, different browsers implement this slightly different depending on design).
All browsers I have come across (that support frames) give you options on which frames you wish to print, if the user does not know how to use their own browsers print options then they deserve to get unexpected results.
Right! They give you an option of "which chunk of this full page do you want to print", as opposed to getting the entire page printed - as you saw it on the website. So again, you agree that the point is valid, you just think that users should be smarter, and if they arent, too bad for them.

Once again, you are entitled to your opinion, we disagree, and I'll advocate the standard that enables more users to use the website.
redmonkey wrote:
Roja wrote: You often can't bookmark any page other than the root - again, depending on browser implementation.
This can be quite useful, having your users entering your site at one single point could be useful especially if that entry point contains your latest news , updates or any important annoucements.
Absolutely. But in a frames design thats your *only* choice. In a standard design, there are any number of ways you could accomplish both - having the user visit a central page first, and then redirecting to the actual content. By using frames, as a designer, you are giving up any choice BUT having no direct bookmark-ability.

Thats now three points where frames give LESS functionality, and css gives MORE - by your own words.
redmonkey wrote:
Roja wrote: Search engines can't index them correctly - leading to poor ratings.
What about the noframes tag?
Compare and contrast - using css, the search engine can catalog my site natively. Using frames, I have to develop A SECOND VERSION of the same content, and put it in the noframes tag. And every time I change the page? Yup, need to update the noframes too.

Twice the work, for the same result. Advantage? CSS over frames.
redmonkey wrote:
Roja wrote: Depending on the design, users might get caught in your frameset, even when clicking on outside sites.
Again, that is down to poor design not the actual use of frames.
No - that is down to users using the native features of their browsers as they are able to on other sites. Frames break the rules of navigation. As a designer, you can always work around that breakage, but it doesnt change the fact that it breaks it.
redmonkey wrote:
Roja wrote: If the browser doesnt support frames, it will literally replace it with nothing - often completely breaking the design. ie, it doesnt 'degrade' gracefully.
What about the noframes tag?
See above. Worse, in browsers that don't support frames (notably some cellphones), the noframes content might not be presented AT ALL! (Disclaimer - I work at a cell company, so I have some knowledge here).
redmonkey wrote:
Roja wrote: So, there are lots of solid reasons not to use them, and few good reasons to do so.
Personally I don't feel that your arguments against frame usage are solid.
You are entitled to your opinion. We do disagree - your arguments amounted to "You can work around that problem" - which doent change the fact that its broken!

Additionally, its not my arguments - you are in actuality arguing against Jacob Nielsen's arguments.. the vast majority of my post is just restating his points.

Posted: Fri Apr 02, 2004 9:35 am
by Roja
LiLpunkSkateR wrote:Roja: like what? (things that aren't possible), i can only think of 1
Unfortunately, my personal website is down at the moment. It will probably be back online early next week. On it, I have a specific design that I couldnt seem to find a way to duplicate in css.

It involved stretching background graphics and similar oddities.. I know that when I tried to accomplish it with css, I failed. Might be that I lacked the knowledge.

If/when it comes back online, I'll post a thread asking for comments on how to redo it using css, and we can work on proving me wrong. :)

Posted: Fri Apr 02, 2004 9:46 am
by redmonkey
Roja wrote:
redmonkey wrote:
Roja wrote: First and foremost, they break fundamental modes of web-browsing. Back/forward buttons don't work consistently with normal web-browsing paradigms.
Having poorly implemented or non existant navigation can lead to problems with users using the browsers back button when viewing a framed website. It is not so much a fault with frames but more so poor design. Relying on the browsers own back button is bad practice/design IMO
So - your opinion is "While you are right, users shouldnt expect the browser to do what they want in my opinion". While the site CAN work around it by having stronger/better navigation, it doesnt change the fact that it is taking away an established paradigm THAT USERS USE OFTEN.

You are entitled to that opinion. But when Aunt Tillie wants to visit your-amazon.com, and cant get around because the back button doesnt work? Your loss, my gain.
Users use the browsers back so often because of poor navigation. In this case Aunt Tillie would use my intuitive and well thoughtout navigation thus her attention would not be drawn away from my site while she directs her mouse pointer away from the page content to locate the browsers back button.
Roja wrote:
redmonkey wrote:
Roja wrote: It makes printing difficult - the default prints the current main frame, not the full frameset (if that even, different browsers implement this slightly different depending on design).
All browsers I have come across (that support frames) give you options on which frames you wish to print, if the user does not know how to use their own browsers print options then they deserve to get unexpected results.
Right! They give you an option of "which chunk of this full page do you want to print", as opposed to getting the entire page printed - as you saw it on the website. So again, you agree that the point is valid, you just think that users should be smarter, and if they arent, too bad for them.


Once again, you are entitled to your opinion, we disagree, and I'll advocate the standard that enables more users to use the website.
Perhaps I should clarify, within these options there is also an option to print all frames. If you want to use a standard that enables the most users to use your website you should stick to ascii only page content.
Roja wrote:
redmonkey wrote:
Roja wrote: You often can't bookmark any page other than the root - again, depending on browser implementation.
This can be quite useful, having your users entering your site at one single point could be useful especially if that entry point contains your latest news , updates or any important annoucements.
Absolutely. But in a frames design thats your *only* choice. In a standard design, there are any number of ways you could accomplish both - having the user visit a central page first, and then redirecting to the actual content. By using frames, as a designer, you are giving up any choice BUT having no direct bookmark-ability.

Thats now three points where frames give LESS functionality, and css gives MORE - by your own words.
At no point have I stated that frames gives less functionality than css, that just seems to be your interpretation. Again the user being able to bookmark a single framed page is down to the user knowing how to use his/her software.
Roja wrote:
redmonkey wrote:
Roja wrote: Search engines can't index them correctly - leading to poor ratings.
What about the noframes tag?
Compare and contrast - using css, the search engine can catalog my site natively. Using frames, I have to develop A SECOND VERSION of the same content, and put it in the noframes tag. And every time I change the page? Yup, need to update the noframes too.


Twice the work, for the same result. Advantage? CSS over frames.
Not quite, for large complex sites you will probably have some sort of contents/index page which contains links to actual content pages. Simple put this link list in the noframes tag of the root page and job done your site will now be indexed by search engines. So yes, I have to update one extra page but if I want to use frames then I think I can cope with that. If you don't have a contents/index page then it's not that hard to come up with one.
Roja wrote:
redmonkey wrote:
Roja wrote: Depending on the design, users might get caught in your frameset, even when clicking on outside sites.
Again, that is down to poor design not the actual use of frames.
No - that is down to users using the native features of their browsers as they are able to on other sites. Frames break the rules of navigation. As a designer, you can always work around that breakage, but it doesnt change the fact that it breaks it.
No, it is poor design i.e. the designer has not specified the target attribute.
Roja wrote:
redmonkey wrote:
Roja wrote: If the browser doesnt support frames, it will literally replace it with nothing - often completely breaking the design. ie, it doesnt 'degrade' gracefully.
What about the noframes tag?
See above. Worse, in browsers that don't support frames (notably some cellphones), the noframes content might not be presented AT ALL! (Disclaimer - I work at a cell company, so I have some knowledge here).
See above.
Roja wrote:
redmonkey wrote:
Roja wrote: So, there are lots of solid reasons not to use them, and few good reasons to do so.
Personally I don't feel that your arguments against frame usage are solid.
You are entitled to your opinion. We do disagree - your arguments amounted to "You can work around that problem" - which doent change the fact that its broken!

Additionally, its not my arguments - you are in actuality arguing against Jacob Nielsen's arguments.. the vast majority of my post is just restating his points.
No, my points were not to suggest workarounds, my points are that with proper implementation there is nothing wrong with using frames.

At the end of day the frames usage arguement is one that is likely to go on until the end of time. From my point of view, the bottom line is I have nothing for or against frames, if I feel they are the best solution for my problem then I will happliy use a frame based layout. If I can use something else which I feel would be more appropriate then I will use something else. What I won't do is simply dismiss frames as useless or deem them something that only a newbie would use.

Posted: Fri Apr 02, 2004 11:15 am
by Roja
redmonkey wrote: Users use the browsers back so often because of poor navigation. In this case Aunt Tillie would use my intuitive and well thoughtout navigation thus her attention would not be drawn away from my site while she directs her mouse pointer away from the page content to locate the browsers back button.
Thats an assumption, and I disagree. Users use the back button because it exists. While I agree that a well-designed site might make Aunt Tilly less interested in using it, the back button is there, is used often, and on a frames site, it wont work. You are still simply providing a work-around for a problem with frames. It doesnt change the fact that the problem is specific, and caused, by frames.
redmonkey wrote: Perhaps I should clarify, within these options there is also an option to print all frames. If you want to use a standard that enables the most users to use your website you should stick to ascii only page content.
Yet again, there is a workaround, but its still a problem. It's a dialog v. no-dialog. It's requiring the user to understand v. the user knowing it will work.
redmonkey wrote: At no point have I stated that frames gives less functionality than css, that just seems to be your interpretation.
You agreed that the back button would not work - and proceeded to give a work around. Thats one.

You agreed that printing used a different modal dialog, one that requires user intervention and knowledge to get the same effect they would get if you were NOT using frames - thats two.

You agreed that directly bookmarking a page didnt work, and cited it as an advantage. While its arguable whether thats an advantage or not, thats still three issues with frames that you agree exist.
redmonkey wrote: Again the user being able to bookmark a single framed page is down to the user knowing how to use his/her software.
Actually, no. You cannot directly bookmark a combination of frames, to get the same end result. That means you have to navigate THROUGH a site to get to the information, instead of going directly to it.

That is NOT the same functionality. Show me how to bookmark a page - with the same presentation (include all relevant surrounding frames). You can't, because the functionality isnt there. Bookmarking is a fundamental paradigm in browsers - it has been there since the 2.0 browsers at least.
redmonkey wrote:
Roja wrote: Twice the work, for the same result. Advantage? CSS over frames.
Not quite, for large complex sites you will probably have some sort of contents/index page which contains links to actual content pages. Simple put this link list in the noframes tag of the root page and job done your site will now be indexed by search engines.
No - the LINKS are being listed. Not the content, not the context, so your score and rating will be significantly lower than if it was the actual content - so I say again, thats less functionality.
redmonkey wrote: So yes, I have to update one extra page but if I want to use frames then I think I can cope with that. If you don't have a contents/index page then it's not that hard to come up with one.
See above. To get the SAME functionality, you have to update the ENTIRE page, not just the links, and EVERY page. You are arguing less impact, but still admit to less functionality.
redmonkey wrote:
Roja wrote: No - that is down to users using the native features of their browsers as they are able to on other sites. Frames break the rules of navigation. As a designer, you can always work around that breakage, but it doesnt change the fact that it breaks it.
No, it is poor design i.e. the designer has not specified the target attribute.
So designers now are required to take an extra step to avoid that shortcoming. I agree.
redmonkey wrote: No, my points were not to suggest workarounds, my points are that with proper implementation there is nothing wrong with using frames.
We are choosing different language. You are saying "Because you can use workarounds, that eliminates the problem".

If there is a rock in the road, and I can swerve to avoid it, that does NOT change the fact that there IS A ROCK IN THE ROAD. There is a rock in the road, I've pointed it out, and you keep claiming you can swerve around it.

I agree. You sure can.

But there IS a rock in the road. Frames DO have problems, and the issues I listed ARE things that are wrong with frames.
redmonkey wrote: At the end of day the frames usage arguement is one that is likely to go on until the end of time. From my point of view, the bottom line is I have nothing for or against frames, if I feel they are the best solution for my problem then I will happliy use a frame based layout. If I can use something else which I feel would be more appropriate then I will use something else. What I won't do is simply dismiss frames as useless or deem them something that only a newbie would use.
I don't dismiss them as useless - I use them on my own personal homepage. Specifically because I couldn't find a solution in css to the problem I faced.

Like you said - I found a situation where they were the right tool.

Doesnt change the fact that using frames brought all of the issues I listed and more. Frames have shortcomings, you agree to that, we just disagree on what the impact and importance is.

I'm all about agreeing to disagree, but please don't claim that workarounds remove the fact that there IS a problem.

Posted: Fri Apr 02, 2004 11:33 am
by redmonkey
Is it a case that frames break browsers or is it a case that browsers that claim to support frames don not fully support them?

Is it the fact that using frames breaks the back button or is it the case that the browser doesn't fully support frames and the back button is poorly implemented?

I don't class any of the methods I have suggested as workarounds for using frames, but I could class them as workarounds for browsers with poor frame support.

Posted: Fri Apr 02, 2004 12:50 pm
by Draco_03
Mac it s true.. but usually 2 thongsfirst it might be on a local machine.. (your phisical server) so who cares about reloading.. or if you use high speed.. who cares about it .. the oonly thing that frames are good for is fpor ppl with slow connection..

or to redriect (adress bar)
..mhh but nyways

your right frame on phpmyadminare good.. but one good exemple doesn t justify all of the "others" bad ones

Posted: Fri Apr 02, 2004 2:09 pm
by redmonkey
Draco_03 wrote: but one good exemple doesn t justify all of the "others" bad ones
That is the problem, as I said previously, personally I don't think there is anything wrong with the frame model but a very high percentage of sites using frames have not taken the time to implement them properly.

That being said, if anyone has tried to emulate frame behaviour using CSS you will quickly find yourself having to use lines like.. voice-family: ""}""; and .... html>body these in themeselves are workarounds for various browsers poor CSS implementation. And as IE does not (as yet) support.... position: fixed; you will not be able to emulate fully a frame based layout using CSS.

Posted: Fri Apr 02, 2004 2:20 pm
by McGruff
There's another F-word on the site: flash.

My Firebird browser is set up not to display flash unless you deliberately click on the item. Without meaning to comment on your own use of flash, I personally find it's very rarely worth the trouble and usually I'm straight out of there.