Rethinking MODx

[Guest]

Welcome, Guest. Please login or register.
Did you miss your activation email?

1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Neutrogeno]

Hi,
these days I'm evaluating the MODx CMS in deep for a forthcoming web project. No doubt it's a great app but, in my opinion, you should rethink some important choices. Here are some guidelines you could follow:

1. Less is more. MODx admin interface should be more clean and organized. Moreover the project website should eliminate any hype or buzzword and focuse on documentation and support. An updated list of MODx powered web sites should be added. Don't forget that the best promotion for a web app is made by its happy users on their blogs.

2. The whole hyerarchy thing is simply old school. Instead you should introduce the concept of "section" and eventually "subsection". Each post should belong to a section (or sub-section). Home page should be a section apart. Friendly URLs should be mapped on the schema section/(subsection)/articletitle. So you could eliminate the archaic and ambiguous folder/file tree at left in the admin panel.

3. Each section should be associated with 2 templates: the first for the section home page, the second for the single entries. Of course two different sections could even share the same templates. Moreover, when posting an article, the user could choice a different template for that particular entry.

4. Template Variables should be tied to one or more sections not to templates. TV are related to data structure not presentation.

5. You should eliminate any embedded visitors logging. It suck!

6. You should eliminate frames (and the ugly file tree at left) in the admin interface.

I hope my suggestion can inspire the excellent dev team. MODx is not yet ready for my forthcoming project but I hope 2007 can be his year.

Neutrogeno

[rthrash]

Did you check out the public beta 5 for 0.9.5? (hopefully w/in 24 hours our RC)

Some of the issues you mention are already addressed, and they're all on the radar. We won't eliminate hierarchy, but we will offer some alternate approaches. Eliminating it would mean less flexibility and choice for our users, which is the antithesis of what MODx is about. With that said, MODx is clearly not the right solution for every project!

As far as promotion goes, we're not really ready to start any big promotion or marketing campaign but I suspect you'll see much broader awareness of MODx in '07 indeed.

[Neutrogeno]

Hi,
I checked the latest beta5-rev1785.zip

As for the hierarchy tree concept, I think with the section/subsection thing (plus the URL remapping) you could offer MORE flexibility to your users. In fact they don't care about the hyerarchical tree at left in the admin panel (it can be misleading: see the current ambiguity about folder/container and files) they are looking for data organization and semantic URLs.

Also the TV thing is in my opinion very important. They are currently tied to templates but in fact they are a concept related to the data structure not its presentation.


(Note: I'm not a programmer but I know many CMSes and Web Framework as user.)

[PaulGregory]

1a. Well, it wasn't a choice to make it less clean and less organized.
1b. There's a lot of support spread around the site, repository, forum, wiki and bugtracker.
1c. There's a forum where people can promote their MODx-powered website; we don't require users to say that it is powered by MODx and many people choose not to.
1d. We have no control over user's blogs. But yes, the best promotion for anything is word of mouth.

2. Section and subsection *is* a hierarchy, and can be achieved using the "archaic" tree. The tree can be hidden and you can view the documents (not necessarily posts or articles) in many ways.

EDIT: And you can use the hierarchy to make your URLs.

3a. I believe there are ways of letting the parent specify the default child template.
3b. It is entirely possible to include a selection of template in the post-article form.

4. There is no such thing as sections; document types are mainly differentiated by what template they use. TVs are related to the type of page, therefore it is appropriate. Template does not solely mean presentation, it covers the data structure also.

5. Visitor logging is easily elminatable in 0.9.2.1; eliminated by default in 0.9.5.

6. 0.9.5 has slightly less frames. The file tree is on the left (as it is in 0.9.2.1) It is not ugly in 0.9.5.

[straty]

Please dont do this
"The whole hyerarchy thing is simply old school. Instead you should introduce the concept of "section" and eventually "subsection". "

It would make it like mambo, I hate mambo with a passion, I mean a wasted a week or so and found it to be totally shit and all the mambo sites look the same. I love the flexability in modx but I`m still a newborn, but hope to have a site up soon.

I have to admit tho that the admin control does look oldschool, but it works. And the documentation is so bad not bad just could be done better for web designers you end up going from post to post to post and once you find a solution (in the forum) its very hard to find the post again. So newbies like myself seem to ask the same sort of questions over and over.

But its

[OpenGeek]

As for the hierarchy tree concept, I think with the section/subsection thing (plus the URL remapping) you could offer MORE flexibility to your users. In fact they don't care about the hyerarchical tree at left in the admin panel (it can be misleading: see the current ambiguity about folder/container and files) they are looking for data organization and semantic URLs.

Also the TV thing is in my opinion very important. They are currently tied to templates but in fact they are a concept related to the data structure not its presentation.

(Note: I'm not a programmer but I know many CMSes and Web Framework as user.)

Neutrogeno, we do appreciate you taking the time to share your thoughts.

The next release following 0.9.5 will feature something called Contexts which will allow you to divide your site into logical sections for just about any purpose, be it for multi-sites, sub-sites/sub-domains, localized content, a specific web-application, or just another logical section of the site content, as you suggest.  A contextual view then will be just as important as the tree/hierarchical view within or without contexts, as both offer important organization benefits to the site definition and the ability to manage it.

Can you explain more what you mean by URL remapping?  This to me, is the part that requires the traditional hierarchical approach to defining a site taxonomy.  It allows those familiar with web concepts to work with their virtual MODx sites in the same way they would work directly with an Apache document directory.  Contexts could then be used to alter the hierarchy to achieve more complex URL mappings (i.e. to URLs specific to localized content, to subdomains of the site, etc.).

Also in the next release, you will be able to attach what I am calling Content Elements (the next generation of TV's), which can be attached to a Document, a Web link, or even another component, such as a Template, Chunk, Snippet, or Plugin.  I believe this will allow users to continue to use them as they would existing TV's, yet still gain an enormous amount of power and flexibility with the new ways of defining relationships between MODx TV's and the other building blocks of the framework.  Obviously, this will mean new manager views and features to accommodate these changes.

[straty]

You can correct me if I`m wrong, I know I`m new to all this but I prefer to use a tv to categorise my content (with a combo box with the categories), for the end user all I do is create a folder and put all the pages inside it, I dont use the folder its just so the end user can see it quickly.

One of the things I found usefull was ditto and dont understand why you need to install it, it should be part of the downloaded package.

I`ll shut up now

[Neutrogeno]

Hi,
thanks for your promptly replies. (The words of OpenGeek are especially appreciated and encouraging )

Regarding the hyerarchical thing, many Textpattern user could explain to you that they like sections. The problem with TXP is that it doesn't have sub-sections; moreover categories and custom fields are only site wide and can't be restricted to a specific section. However friendly URLs in TXP can be arranged as section/category/title to simulate a hyerarchy. The concept of section is the same as weblog in Expression Engine probably the best CMS around (unfortunately it is not an open source project). I share your dislike of flat sites organized only with categories or tags but I think you can get an organized and hyerarchical site without the visual aid of a windows-like folder/file tree.

In my opinion MODx is a little underrated in the CMS world . In fact the application is solid and really powerful. However very few people analyse an app in deep. Most people form their opinion at the first sight and many friends of mine don't like MODx simply for the look and feel of the admin panel (especially in the old version hosted by opensourcecms: please replace it as soona as possible with the new release!). The admin panel can be perceived as overloaded and Windows-like (the culprit is again the hyerarchy tree) by sophisticated people used to live in a world full of clean 37signals-like iterfaces. As you can see I'm a sort of MODx evangelist among my friends 

Neutrogeno

[doze]

@topic
Hierarchical site tree was one of the reason why I choosed this CMS long time ago after first trying out Mambos etc.. I really hate the "flat layout" that many other CMS'es use.

One of the things I found usefull was ditto and dont understand why you need to install it, it should be part of the downloaded package.

It is in next release, it was still under heavy development when 0.9.2 was released..

[PaulGregory]

Right, so this is all about seeing the tree and making assumptions that you have to use it.

So, how about if the left panel had a choice of view?
They could include:
* tree hierarchy
* flat listing, alpha sort
* flat listing, date sort
* flat listing, just my documents
* grouped expandable listing, grouped by template
* grouped expandable listing, status sort
* flat listing, search results

and also the ability to add custom views:
* grouped expandable listing, grouped by TV (and thus enabling categories or tags)?
* grouped expandable listing, grouped by TVs within another TV (and thus enabling section/subsection)?

Manager users could toggle through these, but it would remember the setting for next time.

That way everyone would seem to be happy. It looks like ultimately we'll need to be grouping by context, so this would be a stepping stone to that...

[davidm]

@Jason : I love those nuggets about the future... I love Contexts already

Regarding the hyerarchical thing, many Textpattern user could explain to you that they like sections.

As it happens, I have been a long term Textpattern user, and yes I can tell you what I like about sections. Yet, I don't miss them ever since I switched over to MODx.

Section's benefits are easily replaced with @INHERIT. You can have children document of a folder inherit layout elements for instance

The problem with TXP is that it doesn't have sub-sections; moreover categories and custom fields are only site wide and can't be restricted to a specific section.

And that's exactly why I looked elsewhere. Don't be mistaken, I love Textpattern but it lacks in several areas which make it unconvenient to build corporate websites (which often require a directory like structure, and many depth level, not to mention custom fields).

However friendly URLs in TXP can be arranged as section/category/title to simulate a hyerarchy. The concept of section is the same as weblog in Expression Engine probably the best CMS around (unfortunately it is not an open source project). I share your dislike of flat sites organized only with categories or tags but I think you can get an organized and hyerarchical site without the visual aid of a windows-like folder/file tree.

As it happens, I am also a long term EE user (check my blog )
I would not say it's the best CMS around, it's an hybrid of sort between blog and CMS and just one of the most flexible tool around along with Texptattern, CMS Made Simple and MODx. I had written a comparison between txp, EE, CMS MS and MODx on txp's forums, if you're interrested...

And no, it's not weblog which is the same as section, but template groups
Weblogs in EE are really more like subsites.

In my opinion MODx is a little underrated in the CMS world  

Won't be for long, believe me those who underated it will be surprised by what's coming... and they'll have missed a hell of an exciting ride

Most people form their opinion at the first sight and many friends of mine don't like MODx simply for the look and feel of the admin panel (especially in the old version hosted by opensourcecms: please replace it as soona as possible with the new release!). The admin panel can be perceived as overloaded and Windows-like (the culprit is again the hyerarchy tree) by sophisticated people used to live in a world full of clean 37signals-like iterfaces. As you can see I'm a sort of MODx evangelist among my friends 

I know.

I couldn't agree more that the admin had to evolve, and it has.

There has been a lot of talk and thinking for the future of MODx once we break the ties with the Eto-legacy with the 1.0 "jump" (though there is less an less of Eto in the MODx code, 0.9.5 has made quite a leap already ).

We have enough talent onboard (and coming to us) to bring to you a 37signalish admin when the time comes

[identity]

As a non-programmer (though one who started down the road, long, long ago) and one whose CMS experience is very limited and only limited to Drupal and MODx, I guess you can take these comments at merit.

A way to view files/documents in multiple ways would be great, but I'd be cautious about totally discarding hierarchial ordering... this is just a natural and organized way for most people to view and relate to things.

I have to say that I share some of the views on documentation though. And it should not be easily brushed aside.

Reality, regardless of what you are selling, developing, providing, etc., is that the "best" doesn't always win out. One may succeed in creating the best product around, that has every feature you could ever want, implemented in the very best way possible, and still be beat out by a product half as worthy. In the end, it isn't about features, and in some ways, it isn't about benefits, it's about delivering solutions.... and the initial solution is how to tap into those features to reap the benefits.

MODx is at a very interesting point in time.... lots of development at the core level, and the secondary level (snippets, modules, etc.). Excitement is shared by many about the future, and the present. The team wants to present the next greatest thing, and many posts talk about 1.0, which isn't necessarily a bad thing. Of course, if the team had more resources, time, could spend less time dealing with the same questions over and over again in the forums, 1.0 might come sooner. And because this is in such a development stage, documentation can be hit or miss... in some ways documantation is excellent and I commend the efforts so far.

But part of that reality isn't about what MODx is going to do for anyone tomorrow, it is what MODx can do for you today. And even after a short time working with MODx, I think it can do a lot already. But much of the future, I feel, lies in how well MODx can help people get up and running, understanding, and putting MODx to work. In the script world, no one is too impressed with what will be... I think we've all dealt with scripts that have been promising things for years. Not saying that is the case with MODx, but remember that it is perception, and as MODx grows and gains more attention, much of that perception will be quick and immediate, and quite possibly incorrect, but words spread quickly on perception regardless.

I'm still hoping that the wiki will help address some of this "up and running" issue by serving as a "final and up to date" repository of information. I'm sure many a newbie has already spent countless time reading through forum posts 7, 10, or 20 pages long on this or that, only to find by the end that 1/2 of the issues, hopefully, that had been brought up are now resolved... hoping that we can remember which are which, but still feel no closer to actually understanding how to implement something, and in some cases, may still be trying to figure out just exactly what it is for.

While it may seem like a bit of a waste to put more effor into documentation at this stage, I think it can pay off in the end. It will help more people get up and running faster, some of which may start creating their own solutions to add to the mix if they are capable, others may just help spread the gospel. It will also help free up the countless questions about how to simply implement this or that, which will hopefully free up everyone else to continue moving towards the future. In many cases, hopefully much of that documentation can simply be amended, updated, or otherwise for the future, rather than entirely scraped and pitched.

I think the team has done a great job, and I'm certainly not out to belittle the work or the effort. But I also look at how much time has been spent by myself and my impressions of others just in trying to figure some things out to get a better grasp on things. But mostly, I just want to raise the caution, as we all know, software is a constant "work in progress" and it is all to easy for the documentation aspect to be a perpetual "next release" since so much will be outdated anyway, that it never happens.

And coming full circle, the best apps will be those that provide the quickest and easiest solution, not necessarily those with the best features, and the first solution is understanding and implementation. I just don't want to see MODx not receive the full grace of adoration and respect that I think it deserves... and will deserve.

[rthrash]

I really appreciate the feedback here tremendously, and it is 100% accurate. Hopefully the wiki will help address the documentation gap. With that said though (and I hate to beat a dead horse), but MODx is (at this time) squarely NOT aimed at newbies primarily because of the documentation and some of the pending changes that have been alluded to on the forums. While I appreciate your skepticism over the future, I think you'll have a great opportunity to either say "I told you so" or "oh my gosh!" in 2007. I'll leave it at that.

MODx is for developers and technically inclined individuals. MODx is for web professionals who've got a bit of experience. MODx is not for everyone! If you can open up a snippet, read the source and comments about the parameters, and pretty much figure out what happens within the code there's a good chance MODx makes sense for you. From the pending 095 release announcement:

MODx is ideal for web development companies, technically inclined independent freelancers, and in-house internet departments that need a secure and customizable environment they can confidently turn over to their ultimate end-users to run...

We're definitely more than happy to have "newbies" who're willing to dig in and get dirty and push through a learning curve, in fact we have at least one of those on our core team now! However, we ask for patience and acknowledge things are not as they should be with documentation. That will change though in due course; there's still some legacy baggage to shed before we're ready to start put concerted effort into proper documentation.

[Neutrogeno]

I've just discovered another great plus of MODx: its very active and smart community 

In fact I share the opinion that MODx is ALREADY a solid and powerful app. Instead of adding a ton of new features I would concentrate on 1. documentation; 2. redesign and simplification of the admin panel (use a white background: it inspires cleaness and rationality); 3. massive bug hunting and performance enhancement (especially in large scale websites).

The availability of a stable, powerful and well documented release will give a definitive boost to MODx popularity.

As for the wiki I don't like this form of documentation; I would prefer a traditional online beta book with a table of contents and inline comments. So the documentation on your web site is a good starting point. I especially like the new Django (beta) Book.


@davidm: your comparison between TXP, EE, CMS MS and MODx is really good.

P.S. Suggestion for the manager redesign: free FAMFAMFAM Silk Icons (of course on white background)

[straty]

I`m with you on that one, the features already are great and  with the extra modules/snippets etc you can do most things. Keep it simple, otherwise we end up with a mambo clone. I`m also not a fan of wiki docs, keep it on the site in one place gets my vote.

I'm using the system becuase its clean and simple - end of story