Discussions on the interface

[Parahead]

This is a continued discussion from the Coranto 1.31.2 BETA thread about the Coranto interface.

* Is there a compelling reason to have the style sheet embedded within the cgi rather than referencing a freestanding external file? I think it would be easier for everyone to maintain and modify their style sheets if it were formated and commented.

Yes, I agree it would be better. Changing this would break the current themes and the Chameleon addon.

* There are a couple places where I can't tell where an autogenerated table is originating from. (I think this was in your the XMS addon)

This is one of the problems with fooling around with the layout, some stuff are originating from addons. We might end up with a quite funny looking installation if changing the core but not some of the addons to reflect this as well. Then we have the backwards compatibility issue for addons...

* the autogenerated page bottom navigation is currently output in a format that appends a pipe after each link. The links are compiled from a couple of different sources. I think this would be better if it were redone as an inline list, then assorted CSS styles like those listed here could be easily applied

I agree, but once again, if we change this, the addons that adds a link at the bottom navigation needs to be updated as well.

* It seems to me that the Main index should be displayed like this rather than it's current line up. I don't know if there are structural reasons for the current line up or if anyone besides me even sees this sort of thing as important enough to bother with.

I like your suggested order, just hasn't thought about it. This one is easy to accomplish, I can change it right away...

* I'm wondering if it is structurally possible for the Modify News table to be swapped out with an addon. Then we could offer a scrollable alternative something like these

http://www.imaputz.com/cssStuff/bigFourVersion.html
http://www.imaputz.com/cssStuff/bulletVersion.html

Do you really see this as a big issue? I think it could be achieved, needs more investigation though. Maybe with a Yes/No option if this should be enabled or not?

* I'm wondering if there is any value in adding these Textarea Tools as an option in the CutomFields addon (especially in light of the new GEA addon)

Yes, but this request should go in the addon forum then, not the core forum?

Since you have mentioned being quite busy I haven't directly solicited any help in stomping my self-generated bugs. Assuming that my work will be judged worthy of inclusion and folded into the next iteration, do you (the core coders) want to splice in my formatting changes or should I send you my adjusted files after the new issues are resolved?

Hopefully over the weekend I will get my current version and notes organized and send out a more complete list my recent thoughts.

Yes, please use the weekend and work things over one more time and then send the adjusted files to me and I can take a look what might work and what might not. As I said above, if we do to much changes addons *will* brake. I would like to thank you for your work so far though...

BTW, could you post some screenshots of your current work? Or is it OK if I do it? There are some issues I would like to discuss...

[perlman]

Once again, one minute you complain about no upgrades has been done for Coranto, then next about addons not working when doing upgrades. Quite alot of work is done to make sure addons do keep working when new features are added, but if everything where perfect as it is now, there would be no need to do the changes? I think what frustrates you is the Facelift addon, correct? Well, you see, that addon makes actual *changes* to the core when installed, that is the reason each release of that addon needs to be matched with specific release of Coranto, otherwise you may loose bug corrections and such. That is the reason I said, stay away from that addon.

My thought about this is that we shouldn't need a facelift addon.
For the first, it is quite clear that there is something wrong about the admin section as people always are installing facelift and chameleon in Coranto.
It gives you a clue that changes are needed with the interface.

And like muttly has been saying, the styles should be seperated from the core. I think this is really the issue here. When addons have to edit the core , you end up with trouble and more work.
What I think, would have to be a good solution, is to seperate styles/interface from the core, totally.
Now, this would, as said, make Coranto less backward compatible with addons, but I think it has to be done.

[Parahead]

My thought about this is that we shouldn't need a facelift addon.
For the first, it is quite clear that there is something wrong about the admin section as people always are installing facelift and chameleon in Coranto.
It gives you a clue that changes are needed with the interface.

And like muttly has been saying, the styles should be seperated from the core. I think this is really the issue here. When addons have to edit the core , you end up with trouble and more work.
What I think, would have to be a good solution, is to seperate styles/interface from the core, totally.
Now, this would, as said, make Coranto less backward compatible with addons, but I think it has to be done.

I agree with you perlman, we should not need the Facelift addon. It's just that changing the current css structure to a new way of producing the output isn't as simple as some people think it is. It requires some time to look into the best approach, implement that approach and make sure it doesn't have *too* much impact on backward compatibility and browser limitations. Anybody can have an opinion about how the interface should or should not look, but those people are often not the ones that has to actually make the implementation and make sure everything work.

Personally I feel a little stretched, since I tend to spend to much of my time helping people here in the forum when maybe I should be focusing on developing the core instead. I don't know, to much people asking questions, to few answering...

[perlman]

I know it is hard, and I'm not requesting it. I'm saying it could be like this, but it takes a lot of work and time. But it is an area where Coranto needs to improve.

We should launch a Coranto developer campaign! Trying to get new coders to help out. But unfortunetly, it's not that easy.

[Parahead]

I know it is hard, and I'm not requesting it. I'm saying it could be like this, but it takes a lot of work and time. But it is an area where Coranto needs to improve.

I agree, and it is just one of several areas... The problem is that both this interface issue and the database abstraction layer which is the two important things will probably be hard to implement without breaking some eggs.

We should launch a Coranto developer campaign! Trying to get new coders to help out. But unfortunetly, it's not that easy.

I would say more addon developers and people helping out in the forum.

[perlman]

I agree, and it is just one of several areas... The problem is that both this interface issue and the database abstraction layer which is the two important things will probably be hard to implement without breaking some eggs.

I don't think Coranto can improve without breaking some eggs on the way.
It has to be done. New eggs will come with time, so breaking some of them now isn't such a bad thing.

[Parahead]

I agree, and it is just one of several areas... The problem is that both this interface issue and the database abstraction layer which is the two important things will probably be hard to implement without breaking some eggs.

I don't think Coranto can improve without breaking some eggs on the way.
It has to be done. New eggs will come with time, so breaking some of them now isn't such a bad thing.

We are on the same level about this perlman, no worries...

Although, we have to remember that the person that has taken the legal responsibilty to distribute Coranto is msbzdragn and he needs to agree to certain changes. I know that he praises the backwards compatibilty very high. My highly personal view is that some quite large changes needs to be done, so we do them and increase the version number to 2.0 (to indicate this major change) and "start over" with those core changes (ex. interface and serialization) very improved and more dynamic.

I am starting to get of topic here, and maybe we should start a new thread, feel free to do so if you would like to pick up this particular conversation. The thing is that one reason why there has been quite few core developers involved at any given moment is that the core files are few and if a (larger) change needs to be done more or less all of the core files are involved. This is one of the reasons why a source repository just isn't working for a Coranto devteam. It is hard to work in parallel... I am starting to enjoy that Laphroaig to much after a busy week and should quit this topic before I get to loose...

[Jahandar]

I haven't done much advanced CSS, so correct me if I'm wrong, but it doesn't seem like many addons would be effected by this..? Of course Chameleon and Facelift would, but they would be obsolete anyway. When addons use the methods provided by craddon.pl, that can be made work fine, too. I've rarely seen the style tag used in addons (specifying the width for textboxes is the only time I can remember in the addons I've looked at closely), so shouldn't the CSS be able to accomodate any extra stuff? Perhaps span and div tags might also be used to enclose anything added through certain hooks.

[Parahead]

I haven't done much advanced CSS, so correct me if I'm wrong, but it doesn't seem like many addons would be effected by this..? Of course Chameleon and Facelift would, but they would be obsolete anyway. When addons use the methods provided by craddon.pl, that can be made work fine, too. I've rarely seen the style tag used in addons (specifying the width for textboxes is the only time I've seen it), so shouldn't the CSS be able to accomodate any extra stuff?

I haven't done any advanced CSS either, but when addons print out their own tables (as an example), I guess this will have an impact? Basically the things that *isn't* done through the methods provided by craddon.pl. We also talk about stuff like the bottom list separated with pipes (|) instead of each entry being pushed to an array for Coranto to print out in a custom way. Things like that. Maybe each thing in itself doesn't has a big impact, but put together it may be hard to maintain a proper layout...

[EDIT]I see you changed your original post... Yes, DIV tags may be a solution to one extent and is an approach I will look into.[/EDIT]

[Jahandar]

One other thing... This is kind of off the train of thought we're in, but has been bugging me.

I think it is safe to say that Coranto has fully graduated from a news posting script to a full-fledged contant management system. With that in mind, I think it would be approprate to change Add News, Modify News, Build News, etc., to Add Content, Modify Content, and so on.

Its a minor change, and really an easy one for a user to do even with little programming experience, but I think its important for how Coranto presents itself.

[Parahead]

I think it is safe to say that Coranto has fully graduated from a news posting script to a full-fledged contant management system. With that in mind, I think it would be approprate to change Add News, Modify News, Build News, etc., to Add Content, Modify Content, and so on.

That is true. I have thought about something along these lines myself. Stuff like this (phrasing and use of words) is easy to change *before* massproducing language files, and I agree "News" could be changed to "Content"... It needs to be changed *everywhere* where news is used though.

Back to issue and CSS. I will also look up using unsorted lists (the UL tag) instead of tables. This will (I think?) make it easier to have different layouts, for example with the same HTML but different CSS change a list like:
* item 1
* item 2
* item 3
to somthing like this:
* item * item 2 * item 3

I am just looking at the differences between the normal layout with two columns at the admin page in Coranto and with Facelift enabled the one column "short list". It would be really neat if a new CSS approach could support both of these layouts just by changing a CSS file. I like the subheadings (Coranto Configuration/News-Building Configuration/Addons) at the Facelift main admin page, this is somthing that should be implemented regardless I think.

I really need to dig into the world of CSS to make inteligent decisions about how to proceed and visit sites like CSS Zengarden and ALA to get some inspiration and ideas. If there is some CSS wizard out there that could make *mockups* of the Admin page (for example) with two really different CSS layouts.

Another thing I thought about is the Facelift left menu (or the normal bottom menu), are those the options one would like to see there? Wouldn't it be a good idea to have some setting where you choose which links to present in that left menu (or the normal bottom menu)? Anyway, I guess that menu should be a UL list enclosed in a DIV tag to get full CSS freedom?

[Jahandar]

Yes, I've always liked the Zen Garden site. That is what they seem to do, avoid tables, use ULs and enclose everything in div or span. The difference between the actual HTML and the final page is quite remarkable.

[muttly]

BTW, could you post some screenshots of your current work? Or is it OK if I do it? There are some issues I would like to discuss...

Please do, and Ill follow up with some additional ones. It sounds like you are already thinking some of the same things I am, I just didn't want to be making commitments for you.

[Parahead]

BTW, could you post some screenshots of your current work? Or is it OK if I do it? There are some issues I would like to discuss...

Please do, and Ill follow up with some additional ones. It sounds like you are already thinking some of the same things I am, I just didn't want to be making commitments for you.

OK, here are some screenshots directly from your testsite (URL removed!). I haven't got the time right now to give extensive feedback, just some quick questions I have:
Login Page - Don't think we should use a img/logo
Main Page - Why is the first options not in sync with the rest?
Admin Page - Why are some options "out of sync", not 50% left/right?
Manage Profiles - Rather different colors than the rest of the pages?
Modify News - The most right columns lack headings?
Settings Page - Why isn't the left and right side split 50% for every line?

What I would like to know is if you have used a technique like the one Jahandar and I discussed above, using DIV tags and UL lists instead of tables to produce the output on every page? Because I think that is the approach we should be using...

Also, would it be possible with your current setup to make the lower menubar appear to the left like Facelift does, just by using a different CSS?

Personally I wouldn't have choosen those colors, but of course, the taste differs...

Also, if you (or someone else) could tell me if it would be possible to use the same HTML but different CSS-files to achive the two scenarios below I would be very glad. If you then actually produced a HTML and two CSS-files that worked like that in real life it would be even better!
Two options on top of each other
Two options side by side

[Jahandar]

Okay, I think I've got it...

Both examples use the exact same html, the only difference is the .css file that is called...


Vertical Menu
Vertical CSS


Horizontal Menu
Horizontal CSS

I kept it pretty simple, also I just used tables inside the for the submit and modify li items, but we'll probably end up using divs. It should work either way though.

working on getting the description boxes going, also switching to all divs...