Please vote for: MDL-40160
BUMP!
Please also vote for MDL-372 so posts like this can be pinned if needed.
Anyone knows this commercial plugin:
http://www.inserver.es/blog/urban-alternate-lms-login-3-0/
It provides many features that would be desirable in the moodle login: carousel image background, full configuration with separated settings and theme independant.
I think it could be a good start for a new login renderer
Whilst borderline with this: docs.moodle.org/dev/Moderator_guidelines#Unsolicited_advertising - as its in the context of a reply and in the same topic area of the original post - and possibly a trademark violation - to be honest it does not really help with making it happen in core code as you have to 'purchase' the product first in order to then understand and adapt it. It would be far better for them to contribute their code to the community for the benefit of all.
With 'Carousel' image background, I believe that Elegance already has this. And being publicly available, I ported the code to Shoehorn. So its possible to have a changing image background on the login page.
I believe from a tweet (correct me if I'm wrong) that Martin said at MOOTUKIE along the lines of 'Code and information are free, people are not'. Thus Moodle has more of a 'service' model.
Your eloquence astounds me every now and again!
Nicely said!
Gareth, I don't write so much here but I read a lot from this forum and I really enjoy your comments and I agree with your position especially lately with BS3 for moodle 3. You are one of the few that have a clear idea of the future of Moodle Themes while others still stuck in version 1.9
But my intention was not an advertisement and my apologizes if you understood it was. It was just a comment about how (many) others developers that do not publish in the Moodle plugins directory, design for moodle and have other ideas totally different of the conventional.
Another sample (not commercial) would be the Ulkit theme that has a great potential. It is based in another framework developed for Joomla and has features that BS doesn't includes.
And I could say also exactly the opposite about what you should avoid while creating a login page. You can find a sample in the Elegance theme, which has the worst login page ever created in a moodle theme (fortunately you can switch to the default)
IMHO, taking new ideas from outside moodle and avoiding some errors from the past, we can create better and powerful themes.
The first comment from a client is always the same: I want a moodle site that it doesn't look as a moodle site. And this is always my goal.
Thank you for your explanation Fernando.
Whilst Elegance perhaps does not have the best login page, it was at least an attempt to make one within the constraints of the core code as it was and still is. This is why there needs to be a login renderer. I do understand the need for commercial motivation to keep things private and effectively 'closed source' but Moodle is 'open source'. Anybody purchasing the code just once could make it public under the terms of GPLv3 and if the code is not licensed in that way then its breaking GPLv3 as its a module. Therefore it's better to share and build a reputation as a good service provider for more work than keep things private.
Hi Fernando,
I hope you were not including me in those you thought still stuck in Moodle 1.9! 
I agree a lot with what you are saying, and I suspect that there needs to be some more discussion before we can move on; but generally the main problem I see with Moodle, and have done ever since I have been on this Forum, is the fact that there are lots of areas within Moodle that are badly put together, mainly the HTML.
I suspect the main problem is that the majority of Developers working on Moodle are 'Programmers' and not 'Web-Designers', in that they understand how to create an interactive LMS using SQL and PHP, not to mention JavaScript and AJAX, so their solutions to problems are generally logical, and in some cases dynamic, but not necessarily user friendly or indeed, as is the case for the Login page, not easy to re-style.
Cheers
Mary
No, of course. But I'm sure you remember what happened when new themes structure were adapted for moodle 2 and what happened also when Clean (and bootstrap as well) was selected as a default theme and some people still wanted to keep the "wonderful" old themes in the default installation.
Reading the topic about Bootstrap 3 I remembered those old times and looks like the decision to go ahead will take some time.
I agree with you that moodle still have many problems with the themes structure and that maybe designers should give some advice about how to create a better theme structure. Other OS projects have a designer team that help in the development of themes structure.
I work with many scripts and the most complex to create design is moodle. And personally, I'm a developer and not a designer, it is also difficult to understand for me why moodle is not using another structure less complex and more flexible.
There are many scripts that use simple structures where you can add the content easily to design wonderful and attractive sites and not like the "oh no, another moodle site theme" structure.
The most simple structure I know is "Get Simple" CMS (no, not and ad ) where you need a simple HTML & CSS structure to add the renderers. Really simple.
I just want to notice that others are moving ahead quickly and adapting new technologies while the big Moo is still thinking what to do. Joomla has a similar releases calendar and is going much faster adapting technologies than any other CMS.
As you say, many things should be discussed before move on.
PS: A modern emoticons set would be also great for v3
does voting actually do anything? This is a serious question.
It's been stated before by integrators as a reason for not accepting enhancement solutions, as in 'not enough votes'. In reality I'm starting to loose faith.
Well, we have issues with many community votes that get a cursory look from HQ and nothing more unless someone from the community posts the fix - and then its not always adopted. And then we have issues with no votes at all that are taken on as major improvements by HQ.
So I'd have to say, what effect they have is less and less apparent to anyone outside those actually making the decisions.
Reasons given for non-adoption that I've seen in recent trackers I've looked at include things like 'It's not a complete and perfect solution' - e.g. adding div to the span in the multilang filter. And because its not a complete and perfect solution, the small improvement gets rejected and no change is made for several years as no-one has come up with the perfect fix.
And yet we have Atto???... And still don't have renderers in place for things like the login page which has been requested for ages!
So voting in the tracker - we need to do it, because if an issue has no votes HQ wont look at it at all (unless its one of their own projects): But like Gareth, I am losing faith in whether HQ actually take any notice of those votes. I guess we'll have to wait and see if Martin's Moodle Association carries any more weight, or whether in fact it just takes those decisions even further away from the tracker votes.
Good points Richard.
Perhaps as evidence to back this up is MDL-372 which currently has 55 votes and has been going for twelve years!
With the span / div point, I lost real faith with MDL-37490 which on the surface was relatively straightforward, but after many attempts was put back because of the form code and the patches inability to 'adjust' it even though the changes would have been outside of the remit of the patch. I've given up for a while (if not ever) as it was giving me nightmares.
This is perhaps now why for the most part stick with contributed code as for me I can make timely progress. If I need a change that I cannot get around by any means then I'll submit a patch with the smallest footprint as possible and hope that it gets accepted. Then that will be in the latest dev, then after three weeks can submit a back port request and hopefully that will happen. WOFF2 support is a superb example of this.