Feedback to: Textpattern CMS 4.6.0 beta 3 released

phiw13 · 2016-09-07 13:19:35

maverick wrote #301144:

2) if they are unique and necessary, can we handle them differently than keeping them visible all the time? Maybe that means all preferences become a slide out, non-negotiable preferences a modal, or a warning – “you didn’t set this” pops up.. Just my preference among the various views :)

While re-reading through the various posts in this thread earlier today, I had the idea that Article Status and Sections should be moved out of the sidebar, and into the main content editing block, probably below the main textarea; that leaves the two categories widgets. Those could certainly be inside a collapsible block.

I’m no fan of needlessly adding modal pop up warning. Those are very disturbing.

And the whole sidebar could be itself a show/hide thingie for deep immersive writing.

Hmm I see Bloke had the same idea …

:-)

maverick · 2016-09-07 13:27:56

phiw13 wrote #301152:

I’m no fan of needlessly adding modal pop up warning. Those are very disturbing.

For full disclosure, neither am I. I prefer they be used very judiciously. I was merely trying to prime the well of creative solutions :)

maverick · 2016-09-07 13:50:12

Bloke wrote #301149:

I’ve given this some thought in the past, but it breaks so many content paradigms baked into Txp (convention over configuration being one of them) that it’s a major upheaval for little measurable benefit.

From a conceptual viewpoint, I agree, but I believe it most people would easily adapted. As for the technical upheaval this could result in, clearly you have a better understanding than I could.

Content has to live somewhere.

Yes and no. Conceptually and practically, it does have to live somewhere. If not in the database itself, at least with a reference to its location in the database. However, it does not have to be published.

Ergo my comment on paradigm – content publishing system vs content management system.

While all content need be accessible, it need not be published, especially via an article. We already acknowledge that with links, images, and files.

It is only be because we (out of convention) presume all main items are articles and all articles will be published that we create a necessity for section and status. It’s a publishing paradigm, not a content paradigm.

As much as the publishing paradigm resonates with me – I actually like the publishing status work flow and I appreciate using sections for publishing – I still would like to see our convention evolve from a CPS (content publishing system) to a true CMS (content management system).

Which, yes, probably does involve the things you mentioned like custom fields and custom content types.

Granted, it’d be nice to be able to have chunks of content that could benefit from Textile / markup but are not full articles in themselves. Like Forms, but available to content authors not just designers, that you could “include” when needed.

But would they be useful? Dunno. Proper custom fields will help, but they’re still attached to the article, which has to live somewhere to be displayed in a logical location to the site visitor. Custom content types would be better, imo.

an easy off-the-top of the head example of a useful include: associating a company address.

Suppose you have a hundred employees and five locations. You want to put in the address of the five locations once.

In one section you want to put up employee profiles. You want to associate a given address to the several employees who work at that address.

You also wish to have a section which profiles each location.

Having an entry that lives outside of a section that holds a single address and can then be associated with multiple other entries is extremely useful.

Are there ways to make it work? Sure, but this would be a natural, easy to use, out of the box method.

And as noted, as custom fields are attached to specific articles, so they cannot properly serve this function.

Anyhow. I realize that there is not now, nor in the foreseeable future a desire for this. But I figure it cannot hurt (too much) to speak up now and again. :) edit then again, considering Pete’s comment below, perhaps it can hurt.

Last edited by maverick (2016-09-07 13:59:31)

gaekwad · 2016-09-07 13:50:21

Forgive me if I’m jumping off on the wrong angle, here. Actually, forgive me and correct me. And there’s a risk I’ll end up being completely wrong and looking like a tool, but here goes.

The admin-side UI is a bunch of markup and styling that controls what is read from (and written to) a database, and various files behind the scenes.

It’s a theme and, by its nature, the source code for this theme is available. There’s nothing that I can think of that’s preventing anyone with a grasp of HTML & CSS from forking the theme, tweaking a few bits to their liking, and then using that in place of Hive (or NeutraHiveBlueFlat).

There’s nothing stopping anyone making a totally new admin theme, either from scratch or with one of the myriad CSS frameworks that exist these days. Whether I, you, or Donald Bloody Swain is 100% happy with the layout/scaffolding/etc of Hive is not important, because not everyone’s going to agree, and Hive won’t ever be perfect.

Textgarden, the former home of some Textpattern themes, expired and went pop – it’s now a generic Wordpress review-o-matic website. Bert stepped up and ported some front-side themes from Bootstrap, but the admin-side themes are generally pretty hard-to-find.

You don’t have to use Hive. Maybe it’s time (after 4.6.0 is out, of course) for us as a community to think about alternative admin-side themes. Please don’t break our designer; we only have one, he’s busy and his patience is finite.

Last edited by gaekwad (2016-09-07 13:52:57)

philwareham · 2016-09-07 13:51:50

The status and section in main column are also something I’ve considered. Let’s try it out in 4.7 development cycle.

maverick · 2016-09-07 13:58:22

gaekwad wrote #301160:

You don’t have to use Hive. . . . . Please don’t break our designer; we only have one, he’s busy and his patience is finite.

You are correct Pete. There has been a lot of discussion and back and forth. When one is crazy busy and everyone has an idea how you can do something better, it can get to you. I hope Phil is not discouraged by the back and forth. I’m sure that is not what anyone intends. I think we all view it as a community investment trying to allow community discussion create the best version of Txp. If it has become a stress point for Phil, I will be the first to apologize.

hcgtv · 2016-09-07 14:07:09

gaekwad wrote #301160:

There’s nothing stopping anyone making a totally new admin theme, either from scratch or with one of the myriad CSS frameworks that exist these days.

That’s where I’m heading next, time to get your hands dirty I guess.

You don’t have to use Hive. Maybe it’s time (after 4.6.0 is out, of course) for us as a community to think about alternative admin-side themes.

The admin side has always been a please don’t touch or you’ll break plugins scenario. Not to take this thread into strange territory, but I’d like to explore decoupling the admin from core, and publishing a set of API’s or rules to live by and then let everyone use their own widget set to design a backend.

gaekwad · 2016-09-07 14:10:47

hcgtv wrote #301168:

That’s where I’m heading next, time to get your hands dirty I guess.

Excellent!

Not to take this thread into strange territory […]

(We long-since passed that)

[…] but I’d like to explore decoupling the admin from core, and publishing a set of API’s or rules to live by and then let everyone use their own widget set to design a backend.

Go go go!

Bloke · 2016-09-07 14:25:56

hcgtv wrote #301168:

decoupling the admin from core, and publishing a set of API’s or rules to live by and then let everyone use their own widget set to design a backend.

^{adopting dreamy far-away look} Yes please(ish).

That’s sort of what Sam tried with Escher. The Escher admin side was built with the startlingly familiar Escher-tags (<et:... />) that allowed you to render paragraphs, select lists, checkboxes, and so on, as well as articles, lists, images, pages, partials, yahde yahde. Built with itself, like all good lexers :-)

Txp has txplib_html.php, which is an incomplete library for widget building. Sure, MVC is attractive and “correct”, but I’m doing that sort of thing for a non-Txp project at the moment and I can report that making reusable ‘V’ components fed from ‘C’ontrollers is hard graft. Really hard graft, and messy code spread across hundreds of freaking .html files that are nested to produce something on-screen. Gotta be a better way.

Abstracting what we have in our core Txp library is something I’ve tried (and failed at) before. But it’s the closest thing towards usable page building widgetery that I’ve ever used. And I know, somewhere, somehow, there’s a way to open it up further with some ingenuity and a bit of free time.

hcgtv · 2016-09-07 14:39:10

Bloke wrote #301175:

That’s sort of what Sam tried with Escher.

All bow down to the father of multi site, of which I need after launching all these Textpattern sites of late.

Abstracting what we have in our core Txp library is something I’ve tried (and failed at) before. But it’s the closest thing towards usable page building widgetery that I’ve ever used. And I know, somewhere, somehow, there’s a way to open it up further with some ingenuity and a bit of free time.

I’ve always looked at the admin like the Wizard of Oz, you know it’s back there, and it’ll get you home, but pulling the curtain open reveals surprises.

Imagine a core TXP install with just the rendering engine for pages, no admin to speak of, super secure. Takes multi-site to a whole new level, gotta find Sam ;)

philwareham · 2016-09-07 15:13:08

Just for the record, I’m not pissed off at all. Far from it – I’m just keen to release a new version.

Further UI changes can wait until 4.7dev. I’m happy with it and it works. Sure we can (and will) make changes in 4.7. Not least an image grid view and hopefully a way of inserting images/files directly in write page.

hcgtv · 2016-09-07 15:30:33

philwareham wrote #301178:

Just for the record, I’m not pissed off at all. Far from it – I’m just keen to release a new version.

Damn, then we’re not trying hard enough ;)

I’m keen to get all my sites to 4.6, and then we can start discussing 4.7, as we use 4.6 in production and get more familiar with it.

michaelkpate · 2016-09-07 15:42:33

hcgtv wrote #301168:

Not to take this thread into strange territory, but I’d like to explore decoupling the admin from core, and publishing a set of API’s or rules to live by and then let everyone use their own widget set to design a backend.

Headless Textpattern? I know Drupal and WordPress have done that

So, headless proves hugely beneficial for cross-platform publishing and custom user experiences. It makes publishers, designers and developers happy and, eventually, helps building better products for the public. – Headless and decoupled CMS: the essential guide

Honestly, I think it is mostly for really picky people who want to reinvent the wheel. Not that there is anything wrong with that.

hcgtv · 2016-09-07 15:55:34

michaelkpate wrote #301184:

Honestly, I think it is mostly for really picky people who want to reinvent the wheel. Not that there is anything wrong with that.

And this is where we saw the rise of “the runtime” (I totally stole that word from Kevin Cochrane of Day). By runtime, I mean code on your delivery server which acts on the content to render a Web site. These systems were built to make sense of all the published content assets and generate a dynamic Web site out of it. They didn’t manage the content (that was being done somewhere else, remember), but they delivered it. The runtime enables things like dynamic navigation, personalization, permissions, etc.

Link

philwareham · 2016-09-07 15:59:26

Ok, since we have gone way off topic, I’ll assume beta3 is release-worthy now. Should release very soon now!

Textpattern CMS support forum

#91 2016-09-07 13:19:35

Re: Feedback to: Textpattern CMS 4.6.0 beta 3 released

maverick wrote #301144:

#92 2016-09-07 13:27:56

Re: Feedback to: Textpattern CMS 4.6.0 beta 3 released

phiw13 wrote #301152:

#93 2016-09-07 13:50:12

Re: Feedback to: Textpattern CMS 4.6.0 beta 3 released

Bloke wrote #301149:

#94 2016-09-07 13:50:21

Re: Feedback to: Textpattern CMS 4.6.0 beta 3 released

#95 2016-09-07 13:51:50

Re: Feedback to: Textpattern CMS 4.6.0 beta 3 released

#96 2016-09-07 13:58:22

Re: Feedback to: Textpattern CMS 4.6.0 beta 3 released

gaekwad wrote #301160:

#97 2016-09-07 14:07:09

Re: Feedback to: Textpattern CMS 4.6.0 beta 3 released

gaekwad wrote #301160:

#98 2016-09-07 14:10:47

Re: Feedback to: Textpattern CMS 4.6.0 beta 3 released

hcgtv wrote #301168:

#99 2016-09-07 14:25:56

Re: Feedback to: Textpattern CMS 4.6.0 beta 3 released

hcgtv wrote #301168:

#100 2016-09-07 14:39:10

Re: Feedback to: Textpattern CMS 4.6.0 beta 3 released

Bloke wrote #301175:

#101 2016-09-07 15:13:08

Re: Feedback to: Textpattern CMS 4.6.0 beta 3 released

#102 2016-09-07 15:30:33

Re: Feedback to: Textpattern CMS 4.6.0 beta 3 released

philwareham wrote #301178:

#103 2016-09-07 15:42:33

Re: Feedback to: Textpattern CMS 4.6.0 beta 3 released

hcgtv wrote #301168:

#104 2016-09-07 15:55:34

Re: Feedback to: Textpattern CMS 4.6.0 beta 3 released

michaelkpate wrote #301184:

#105 2016-09-07 15:59:26

Re: Feedback to: Textpattern CMS 4.6.0 beta 3 released

Board footer