[TYPO3-hci] TYPO3 4.2 blog entry

[Erik Svendsen]

Hello Dmitry Dulepov [typo3],

> HansPS wrote:
> 
>> http://impossiblearts.com/blog/2008/04/30/typo3-42-usability/
>> 
>> just for your infomation
>> 
> Well, this guys says exactly what I say for two years already on each
> meething with any TYPO3 developer...
> 
> He is right. But no one will change TYPO3 Backend so much. It requires
> huge development and no one will sponsor it. I would say it will take
> at least half a year full time to make a completely new Backend, which
> keeps all TYPO3 power but makes it in a simple friendly and effective
> way.

You, Hans and Hannes isn't the first one who said it. And Dmitry is correct 
about it is a really huge development to really change the backend. But I 
also think it in a more easy way could be made better.

But I think there also is another point thats importantant regarding to slow 
development of the HCI in the backend. A lot of TYPO3 installations from 
companies which are working really much with TYPO3 are often well configured 
for Editor needs. Not like consultancies which maybe do 1 - 4 cheap installations 
and Hannes refer to. I don't think any of my customer has more than 4 - 7 
icons in left menu, and probably every element/choice they don't need are 
well hidden away. And they are configured with shortcuts and actions. A lot 
of these installations are working well for the editors, and then the usability 
push from the customers are more and less missing. 

Else, I think Hannes Moser miss some important elements, but also has the 
idea about one of the solutions. A solutions which a lot of people already 
has put on the table in this tread. What he miss is that you should never 
let an editor have admin right and see the whole admin interface. Of course 
to many do that. But he has the correct idea for solution I think, an kind 
of easy setup which you could choose ons install, or to say a kind of "Quickstart" 
in a new package. Read this list for year 2006, there are a lot of really 
good suggestions, and also some stuff from Jens Hoffmann who is worth reading.

I also find something I will call a misunderstanding in the Hannes blog, 
he compare TYPO3 primarly with Wordpress (which also are my prefered system 
for weblogs, but not for feature rich websites). If you are going to discuss 
usabilty from editors point of view, you also has to compare against really 
feature rich ECMS as eZ Publish, Alfresco (and more), and look how they solves 
it. Because for a lot of potential users of TYPO3 these and commercial software 
are the alternatives.

I also feel like he don't really understand the logic behind TYPO3 and the 
possibilities. Do you have to make a new page or category to write a news 
article. Of course if you want to have a new page. But if you want a new 
content element on a page or a new newsarticle, the editor don't need to 
make a page or a category. 

To give some short comments on his suggestions.
* Easiest-to-use TYPO3-setup AKA "new Quickstart". Yes - and some of the 
work of WEC and YAML are good startpoint.
* Easier asset-managment. Yes, this is the part my clients mostly are complaining 
about.
* Improve the speed of th UI. Haven't tested enough, and as far as I know 
are libraries used for the Javscript part.
* Reduce the options for user. Of course, but are easy to achieve already 
today. But there should be a kind of chooseable standard setup for backend 
user.
* Tabbed menu should never be needed. Yes and no. Sometimes it's correct, 
other times it should be possible to turn it of on BE users level.
* Increase size of fonts. At least the should have been easy to change size 
on without breaking the layout.
* Redesign menu/sites. Maybe or maybe not.
* Bigger buttons. For editors yes, for myself as a person who implement and 
have the administration of sites, no.
* Right use of AJAX, yes, heavy use of Ajax, only if it makes usability better, 
which in my opinion not always are the case.
* Animation: Be really careful.
* Contect Sensitive help: Yes.
* and the last one I didn't understand, because it looks like Hannes has 
misunderstood something.

Personally I think there are three (maybe four) improvments which can be 
made, and who not need a half year of work.
A. Make a new "quicstart" package which gives an easy setup but still configurable.
B. Make som predefined user setup, chooseable when making a new user/usergroup.
C. Make the backend more skinable, also because this will give more alternatives 
and development of better backend.
D. Rename actions and shortcuts and make it possible with them to set up 
editorbackend which are really one-click backends.

All of them are old ideas, 1 - 3 years old.

WBR,
Erik Svendsen
www.linnearad.no