Date: prev next · Thread: first prev next last
2013 Archives by date, by thread · List index

Hello Michel,

Le Fri, 06 Dec 2013 01:09:37 +0100,
Michel Renon <> a écrit :

Hi Charles,

(Sorry for long delay in my answer, but busy in my daily work)
Thanks for taking time to answer.

I think that I should have introduced myself, in order to better
explain my arguments :
I'm a mechanical engineer, software engineer and teacher (in a 
professional way : I have diplomas for all of them).
I'm writing software since 1984, when I was teenager. Ok, it was only 
bad basic, on an Oric Atmos (!). Then I bought my first Mac in 1987
and I started to write scientific software form my mechanical
studies. Then, finally, I could learn how to write code
professionally. Since 1992, I wrote mostly scientific software
(mechanical, also for medical), databases, then web sites, and for
last 5 years I work with OpenERP.
Since 1992, I listen to customer's needs and I have to translate that
in code. And given my customer's feedback, I have some skills in
"listening to users".
And since 1987, my goal as a dev is to write software that is really 
easy to use, so since 1987, I've done UX design.
My biggest project was a mechanical software for the leading european 
aircraft manufacturer, and 12 years after, it's still a success for
users. And another example few weeks ago : the customer wanted a new 
functionality, and other dev started something complicated. I just 
discussed with people, asking few questions and then identified that 
they need not one but three answers. One answer for every context,
every use case. Last but not least, every answer will be very easy to
code ! So I'm not a newbie when I talk about UX Design or "listening
to users".


(all my answers are inline)

Le 25/10/2013 17:08, Charles-H. Schulz a écrit :
Hello Michel,

I feel I need to address some of your points without talking too
much about the StartCenter proposal (which by the way I like very

Comments inline.

Le Fri, 25 Oct 2013 16:37:03 +0200,
Michel Renon <> a écrit :

First, I just want to say that what happens for the start center
is a huge fiasco for LibreOffice. It clearly shows that there is
no design process :
- a subject is thrown for GSOC without any initial work, studies,
validated UX design, prototype, nothing. Just "It would be useful
to present few recently used documents there (as thumbnails), and
do more fancy stuff." [1] And all others GSOC subjects (with UI/UX
part) are the same.

It seems you are expecting a professional process of people actually
working on a regular basis for that. I don't think that's the case
here, and it does not change all the skills and talent the team has.

- a student starts to code, without any information about UI/UX

- the student then wants to enhance UI/UX and makes some completely
improvised assumptions about what users expect [2]. In this thread,
you can read that Mirek starts a design whiteboard at the end of
July, half time of GSOC.

We must at least give credit to the Design/UX team for having
enforced a proper whiteboards-based process.

Just compare them with one test I made :

status, schedule, use cases, technical context, metric, measure for
each proposal...
All of these elements are missing in the whiteboards while they are 
minimum in any professional design process.

So my question is that how it was possible, with all your expertise,
not to include your process guidance and subprocesses in the design
process as it is documented today.

Again: this is a community, you
don't expect an homogeneous team to pop and start working out of the

Devs have created a state-of-the-art process to build LO :
repositories, code review, validations, automatic builds, automatic
tests and so on... Why can't it be extended to the design ?

IMHO, it indicates that UX design and graphics are considered 
second-class citizen in open source projects : they are not directly 
linked to the code, they are here just to make the software look
nice, so they are not very important.
Even Mirek (Design lead) has difficulties to have is proposals
implemented !

Yes, and while it is not ideal, that's where community processes (not
design ones) enter into place to ensure that contributions from the
design team (or other teams) are taken into account, not necessarily in
an automatic way, but  are at least seriously evaluated. 

My opinion is that the design process should be integrated in the
global building process, just like other type of products (cars,
buildings, any manufactured product...)

But your opinion is wrong, because you think a Free and Open Source
Software project works "just like other type of products". It  does
not, never will, never has. 

- he codes what he wants (because there is no roadmap, no

Yup, on the other it's called software freedom....

So LibreOffice is the *only* software/project that has no roadmap !...

the *only* ? please do list the FOSS projects that *have* a roadmap.
I'm speaking about community led Free and Open Source Software projects,
not corporate projects. It's very different. 

it looks like a joke !
I'm sure people from MIMO and other professional users will
appreciate that information.

Oh, the people from MIMO know about it, and don't shake the MIMO flag
like a threat. I met them several times a year and they're familiar
with it. As for the joke, Free and Open Source Software IS a joke. The
Linux Kernel is a joke: it has NO  roadmap! 

Please do get familiar with how Free and Open Source Software work.
Here's a good start: 
(in French). You will see that we are crazy, that we are fools, and
that somehow it works. 

- at the end of GSOC, the start center is unfinished, the UI/UX
brings lot of regressions. Worst of all : some huge lacks in a11y.

Do we
- use the proposal Mateusz gave us?

While being a very good proposal, it is based on broken

Why are they broken?

no real studies : no goals, no roadmap, no use cases, no metrics...

However, it shows that a skilled/professionnal designer can quickly
propose very interesting mockups, even if the designer seems to be
a graphist (not a UX designer).

True. But surely we should not burn the said graphist because he's
not an UX designer?

Please read my blog entries about this subject : 
(paragraph "No UX skills/knowledge")

tl;dr : UX and UI (graphics) require completely different skills.
It's a common mistake to think they can be handled by the same person
("a designer") : he just has to draw a screenshot, or change some
icons, some colors (cf dribbble or devianart).
I'm not the only one to fight against that :

So have you contributed a different proposal? 

If you know about an UX designer who could contribute
his/her time here, please bring him/her here.

To be fair, I think that a professional designer won't waste time in
an unstructured team.
Before contributing, a pro designer would ask simple questions :
- who are your users ?
- what do they expect ? what are their problems ?
- what are your goals ? your vision for future versions ?
- what is your roadmap ? your technical constraints ?

As you don't want to /can't answer most of them, I can't imagine any
pro designer will contribute.

On the other hand, and regardless of whether your analysis here is true
or not, we're not going to change the way we are and how we work just
in the hope we might get (unpaid) professional designer who might
become interested in what we do. 

A designer's time is as precious as a developer's time.
so when Mickael Meeks wrote:
"Designers should lead by inspiration, good relations with
        developers, and producing designs so compelling that
        developers cannot resist taking time to implement them"

it's a clear misunderstanding of design process : designing takes a
lot of time and energy, and the proposal would be thrown by someone
that just says "I don't like that so I won't code it"...

Another example : the color picker.
How many hours spent in designing that ? (whenever it's completed or
not) Will it be implemented any day ?
You may now understand why people don't rush in to make proposals.

I certainly know it can be very frustrating for the design team. But
again, aside community processes, I don't see what would work. 


     - clearly define a UI/UX design (with prototypes), validated
by every team

No. It needs validation by the development team for feasibility,
marketing; perhaps NLC if they find something problematic. But
asking each and everyone will not work because you will never reach
100% agreement. And FWIW I had my share of disagreements with this

I was not precise enough : I meant one person of each team.
The complete process is described in my blog : 

Why not. How come you have not engaged with people here with your
proposal? I might have missed it though. 

     - only then, start to implement it

It is a standard process when you want to build anything (a
physical product, a building...).

We don't build a product, we build a community. If you want to
build a product, please talk to MS Office, they're pretty good at

Uh oh... I absolutely do not agree :

TDF and Microsoft have the same kind of product : an office suite.
But they use completely different paths to create those products :
- MS is a private company, with values of Intellectual Property,
closed software
- TDF is a foundation, with values of Free Software

What is important is :
from an user point of view, MS and TDF are similar : they both create
an office suite. And users just want an office suite that works well
and that fulfills theirs needs (in a professional context or not).
And the idea of "building a community" is something internal for
TDF : it's the way TDF is internally organized to create the product.

So LibreOffice is your product. And building a community is an
internal process.

There's nothing internal about that, i the case of a Free Software
project. Haven't you followed that we are not marketing nor doing a
product? The community is what we do. And what you say about MS Office
and LibreOffice being similar for a user does not change it. I believe
people should be free to be "consumers" with LibreOffice, but that the
values we push forward may be communicated to them, and in doing so,
attract them to our community. So in this regard, it's very different,
regardless of a corporate use or a home/private/small business use.
We're not interested in gaining customers. We're interested in more
users and a broader community. 

The way LibreOffice is developed today is the best way to shoot
himself in the foot :
- don't listen to users (is it the Gnome way of doing ? [4])

I don't think so, but I assume you know Henry Ford's famous
sentence: "if I were to listen to my customers I'd still be selling
horse carriages".

Is this sentence your only argument ?
Mine is based on 20 years of listening to users :
we *always* have to listen to users.

No it's not my only argument, far from that. But listening to users
often leads you nowhere. Just a few examples:
- users want total macro interoperability.
- users want total document interoperability.
- users don't want to change
- users want something nice and beautiful
- users want everything for free
- users don't care about free software
- users want a mac.

All these above are true even if some are contraddicting some others. It
may not be true for all of them, not for all populations, etc. I wish
you good luck to handle all these at the same time. If you spent 20
years listening to them, then perhaps you should know that there can be
different solutions for each of these categories, but that LibreOffice,
being a community led software development project, is only partially
driven by its users. It is however driven to a large extent by its
contributors. That's where listening to users does not work. I'm not
saying it's useless. I'm not saying it's not worth it, or useful, or
even interesting.  I'm saying that the users' input will not drive the
design and the development of LibreOffice. How we can be that popular
with this attitude may suggest that we might be on to something good :-)

And we have to be very careful when analyzing what they said :
sometime we can use directly their requests, but most of the time, we
have to translate, reorganize, mix, split, ask other questions,
analyze and analyze again to guess their real need and find the best
solution. But don't forget that the solution is an answer to their
problem. Not what we want to do.
So we have to listen to them.

KDE devs now also work like that :
   "When designing an application for a user, we should focus at
his/her expectations (which are not necessarily ours)."

And another recent article :
"Listen, observe, iterate
You shouldn't really stop listening to your users, but the point is
you aren't always going to hear from them exactly what they need."

So, yes, it's not an easy job, but it has to be done.

Again, have you made your proposal to the list when the existing design
process was discussed and set up?

- no roadmap (Charles clearly said that [5])

Yes. This is a Free Software Project. Please learn more on how we

- incoherent UI/UX [6][7]

So the process is not perfect.

- schizophrenic behavior in design team [8]

I disagree.

- ship unfinished/undesigned features (template manager for LO4.1,
and today the start center)

We - and specifically this team - have a resources problem,
thankfully not one of talent. Are you a designer? Maybe you could

I already helped : see below.

All this make me feel desperate about LibreOffice. Really. And the
facts are here :

It is the end of the world as we know it and since Forrester says
so it must be true. But what do you do of *all the other articles
about LibreOffice*  :-)

As most users start switching to online and mobile version, there
is no free office suite to compete GDocs or Office365 or Office for
iOS/android. (where is LibreOfficeOnLine ? it would have been one

Where is your contribution?

Just read my wiki page !

Seriously, I can't stand this attitude:
negativity and nothing else. If you want to change and improve
things, go ahead.

I done it :
- a color picker prototype : 
   *I was the first and still the only one to do something like that*
And it was ignored by the design team : [19:18]
This is really shocking because prototypes are foundations of design 

- proposals on my wiki page
   specifically, I analyzed the state of Impress 4 :
   and made several proposals :
     - urgent :
(my mistake was to ask feedback on mailing list instead of creating
asap bug report)
     - short-term :

   Then I made a proposal about entrance animations for Impress.
   You can see that I followed the process I was talking in my blog
   but that proposal was refused by the design lead without any 
valid/scientific arguments :

   Then I gave the results of some brainstorming about a future UI :

- articles in my blog : a constructive critic :

Are all my proposals "nothing else" ?

Just one remarks and one question:
- I see your proposals are either on your wiki page (your user page)
  or outside the wiki, and outside LibreOffice's infrastructure. You
  realize that for someone like me (and please tell me if I'm  wrong) I
  have never seen any of your contributions before. Have you discussed
  that with the design team? 
- I find it hard to believe your proposal was ignored, but again, not
  having been part of it, let me ask Mirek or Alexander: was this
  proposal discussed?

If you are here to complain that life is tough and that we
don't do our jobs right, then go back to your silence, which is not
a solution because you won't change anything that way.

Everything I created for LibreOffice is still available in the wiki,
in the mailing list archives. You can use it, if you want, well, if
devs want. You can't tell me I done nothing.

Indeed. I find you quite shy in fact, and I'm serious. To me it does
not look like you were ignored, rather that no one knows about it.
Which is a problem since the start. 

Given your last blog entry 
its clear that we completely disagree about designing software, so
it's better for me to stop making proposals for LibreOffice for now.
And that last blog entry sadly confirmed that LibreOffice is made by 
geeks for geeks only.

What would be better is that if your proposals were communicated. I've
never seen your name, never read any of your mails (except this thread)
before. I don't read everything on the design list, but  I believe that
if you have not engaged with the design team first, nothing will happen.


Charles-H. Schulz 
Co-founder, The Document Foundation,
Kurfürstendamm 188, 10707 Berlin
Gemeinnützige rechtsfähige Stiftung des bürgerlichen Rechts
Legal details:
Mobile Number: +33 (0)6 98 65 54 24.

To unsubscribe e-mail to:
Posting guidelines + more:
List archive:
All messages sent to this list will be publicly archived and cannot be deleted


Privacy Policy | Impressum (Legal Info) | Copyright information: Unless otherwise specified, all text and images on this website are licensed under the Creative Commons Attribution-Share Alike 3.0 License. This does not include the source code of LibreOffice, which is licensed under the Mozilla Public License (MPLv2). "LibreOffice" and "The Document Foundation" are registered trademarks of their corresponding registered owners or are in actual use as trademarks in one or more countries. Their respective logos and icons are also subject to international copyright laws. Use thereof is explained in our trademark policy.