Which is an easy task compared to getting over mb since... well..



... ... he's not in California. ha! averted!

It's not my problem if you refuse to deny the facts like that and live in your happy little dream world.

Don't worry, you'd be the first, and one I wouldn't regret!

I give you the credit of talking me into using it to. I've already built 2 sites in it, with fully functional AdminCPs and content mangement.

Wow, what an ***. Don't worry about it, I'll save you the trouble.

You won't be seeing me again. I won't regret it either.

bye!

be careful not to trip over someone who cares on your way out

Did Hellequin just leave Massassi? Did something happen that made him a hated member of Massassi while I was away?

It's called a divorce, Zloc. Sometimes mommy and daddy need some "alone time."

Hombre, I thought you would be with us forever. Are the words you spoke that one night under the moonlight all false?

SHUT UP! Daddy's still out getting the groceries! Just you see! One day when he comes back he'll show you ALL!

It's the programming model that's ":awesome:". You still need someone to come up with the HTML for it to spit out. Also, I'm not sure whether the MVC site is actually made in ASP.NET MVC since it's only one part of the ASP.NET site.

I doubt the ASP.NET MVC site is actually using ASP.NET MVC. Stackoverflow is a good example of a nice ASP.NET MVC site, or so I've heard.

WTF? Using an MVC framework as opposed to any other architecture really doesn't impact anything other than the code. You can't have a "good example of a nice ASP.NET MVC site" unless you're talking about the source code. It doesn't make sense to point to a website as an a good example of using a particular framework.

Also, the point is somewhat irrelevant in this thread since CM has already said he's not using C#.

[Just to avoid confusion, I actually really like ASP.NET MVC framework, I'm not arguing against its use.]

Yeah I'm finding CodeIgniter's MVC framework to be pretty nice anyway, so the MVC aspect of ASP.net doesn't even matter here. And I've already made my decision and started programming here so you can stop arguing over ASP.net now :p

The web framework does make a difference as far as the user-facing interface goes, the fact that it's MVC is irrelevant - MVC web frameworks don't JUST do MVC (if they did they wouldn't be web frameworks).

For example Tapestry is a Java MVC framework that gives you terrible urls (which you can't do anything about), terrible html (which you can do something about if you put serious time and effort in) and severely impedes any attempts to produce a site that's any good. It encourages using links to do POST-style actions for example. Then you have frameworks like Django and Ruby on Rails which are ultimately down to personal preference because they don't force you to make sites that are fundamentally flawed in structure and do a pretty good job of encouraging best practices.

This is why the choice of framework matters, and when I looked at the html for the ASP.NET MVC site, I saw examples of things that were fundamentally wrong - which is why I was unimpressed.

Incidentally, the reason why MVC is a good architecture choice for the web is because it closely mirrors the URL->Entity Request + Method->Action nature of HTTP.

I hated the lack of template support in CI but I wrote a library for templates just to get in the hang of it. What did you do about that?

I don't see the need for templates, and CI actually does have template support, it's just optional. http://codeigniter.com/user_guide/libraries/parser.html

I already break up the views as much as I can anyway (headers, footers, menus, etc), and avoid code in the views where possible, so the benefits of templates are few at that point.

template inheritance is the best thing ever, I'll never go back

I am. It's closed source at the moment, but Jeff Atwood is working on it so it's probably good.

I'm not sure Jeff Atwood doing something means it's going to be good, he's spewed a lot of pretty poorly reasoned posts over the years.

he's been pretty clear that he doesn't even consider himself a very good programmer

but he's probably yelling enough at the other developers to make them work better

Well the parser wasn't enough, they needed the ability to act as their own view so you could parse PHP and pass parameters

Rob has a good point. One of the main problems with the Massassi Steam group is that nobody knows who anyone is because our forum handles are different than our Steam names. It would be helpful if there was a link to our Steam ID on the left-hand forum name area thing, or even just a database of our Steam names or something.

Expand that into other popular things like PSN, XBox Live, Wii?

...isn't that what the game forum is for?

The game forum gets used a lot.

Firstly, as others have mentioned, you really should stop treating Jeff Atwood as your personal coding god. Read his blog and think about stuff he raises, but NEVER treat what he says as some form of gospel (incidentally, this is omething he continually points out as well).

Secondly, if it's closed source, exactly how is it supposed to be an example of source code?

I barely read his blog.

I meant because he's involved in it (i.e management).

It's not an example of source, it's an example of a well functioning site with semantic markup.

I would suggest that the involvement of Joel Spolsky is a lot more impressive, since he's a former Microsoft program manager - basically, a guy who can program so well that it's more profitable for him not to.

DO IT



When I made Geb a Never-Ending Story Comic site PHP thing, I mixed PHP and HTML together in one large monstrosity :suicide:. I will never do that again, templates all the way.

I bring it up because wasn't it YOU who edited it when Geb changed the site design? You must have wanted to once you saw my code. >_>

[Edit: I hate when I do this, I used to be able to delete these but nooo...]

I did, but that's not really templating. One of the things with MVC in the first place is that the View contains the HTML. The controller handles the requests, and the model does the "work" on the data. The only PHP found in my Views is nearly identical to what you'd find on a templating system (if statements, some loops, and lots of echos). I don't see the point (as CodeIgniter states themselves, see "CodeIgniter Does Not Require a Template Engine"), when the templates you use are basically synonymous with the PHP you are replacing it with. Why bother learning another language, the template, when the PHP is just about as easy to use and will run faster?

vBulletin uses templates. Their control structures are abstracted out to things like <if> </if>. To me, this does nothing but add complexity, as now the PHP must scan the document to find all the variables and control structures you've placed in it, replace it with the fitting PHP, and then execute it. On top of that, an IDE isn't going to be able to differentiate the control structures from the HTML (except maybe complaining that it has no idea what the hell an <if> is). It just makes it harder to read. I'd much rather use the alternative syntax of PHP (As CodeIgniter documents themselves as a recommendation), and save myself the trouble.

Make sense?

How can you not use templates with CI? I mean, does each of your view files contain your HTML headers and footers? So if you have to make a change to one you have to make a change to them all. Not very efficient. Or are you using the stock parser?

The parser is not excetable solution because it does not allow you to pass an array in the template files like you can the view files. Well at least it wasn't excetable in my case. I created a really simple solution to this, a custom template library.

It simply loads my template view files (header/footer), at this point I use the parser the replace variables with PHP variables, like {url}, {page_title} or {header_tag_insert} and then it will place my controller view file right in the middle.

So in my controller, rather than calling load->view, I call template->show('view',$passingarray1, array('page_title' => 'Page title')); (after auto loading my library of course) So I don't have to use the template class if I don't want to.

Well here..

libraries/Template.php

No, I don't. I use multiple views.

I have a header view, a menu view, a userbar view, the various content views for each controller, and a footer view.

I throw into this into a view model I've created which takes the content views from the controller, and its data, and passes it off to the views as necessary, while loading them in their proper order (or skipping a view if necessary, such as redirects not requiring a userbar or menu).

I see, sounds similar

So I'm trying to wrap my head around this whole project here. As far as the projects/developers 'profiles' go, will it look something like imdb.com profiles? Say you search for a user (actor) it brings up their 'profile' and listed is information about the user, projects they have worked on, and are currently working on, with links to each respectively. Am I grasping it correctly?

I think it sounds good. Someone mentioned earlier on that focused websites or communities do better than general 'we do it all' communities because the groups for each particular engine/code/program/whatever are smaller. If we support many formats of 'development' are we going to give each format a seperate section of its own with info/tutorials/forums ect... or are we going to have to browse through 200 forums to find the section that relates to our particular interest?

Here's an example of what I want to avoid. The structure of these forums I visit is horrible IMO.

http://www.8thcivic.com/forums/

As the site grows we will expand the forums, but all of the editing style forums will be in their own subforum, which will allow you to drill down to the specific "engine" you're looking to work on (Source, UT3, Quake, etc). We won't have many (I'm aiming for 5 or less) additional forums to start, and will expand later.

Tutorials/maps/models, the whole shebang, will be in a full category, regardless of engine/game they are designed for on the main site. In order to assist in finding materials for specific games/engines, the site will make extensive use of a search and filter system. The reason I am not making separate categories for each game on the main page is we will end up with a rather ridiculously large list of categories as time goes on.

Using a tagging system, material of any kind can easily be found, and you will be able to "subscribe" to tags for easy viewing later, as well as RSS feed updates.

As for the user profiles, it will have everything that user has done in the content area of their profile. Tabs will separate based on models/maps/textures/tutorials/etc. You will also be able to subscribe to a user or get an RSS feed for anything the user does.

Projects are a special case, which while falling under general content, will appear more like user profiles, with all the same functionality.

Sounds good. I like the 5 or less idea. Keeps the community broad, but not scattered. What will be done about the dropping/adding of old/future games? Say we have 5 platforms: Source, UT3, Quake, CryEngine, Other. One of them is replaced by the new engine, will the older section be removed or stored in an archive? I know you plan to keep the files, but will the community remain until there is virtually no further interest?

For the content, those sections will never disappear. They'll drop down to the bottom of the filter list (probably in an "Archive" breakdown), but they will still be there completely functional, and people can continue to add to them. Content areas will never be replaced by other ones. It will merely be added to. That is why I don't want to do site sections for each engine.

As for forums, forums that stop getting any interest (And I mean like, less than 1 thread a week or something) will be moved to a closed archive subcategory, where they will be collapsed by default.