Date: prev next · Thread: first prev next last
2022 Archives by date, by thread · List index

Hi Jan, all,

sorry for the late reply. Honestly, I was expecting someone else's
(Thorsten and Emiliano in first place, the current and upcoming board in
second place, the staff in third place and other community members in
fourth place) reply. But nothing came in so I assume it's on me to
re-reply here.

Il 22/12/21 23:28, Jan Holesovsky ha scritto:
Hi Marco,

Thank you for your questions!

There has been a lot of positive changes regarding the Online in the
last year; like that the CODE docker images have no limits of users or
documents any more; that the documentation is freely available to
anyone at ;
It was clear to me that some more extensive documentation regarding
online had to be written, especially on the API part. I can acknowledge
you did a good job.
that a support channel for CODE has been set up:

and that we (Collabora) have started actively supporting the
LibreOffice Technology brand in Collabora Online:
Okay, this trademarks are not present in the version I built on my own a
while ago but I guess has been added later on or is simply available
only in the paid version.
and on our webpages:

Regarding your concerns - please read inline:

Marco Marinello píše v Po 20. 12. 2021 v 20:34 +0100:

I have already said this many times but I want to repeat it: it has
to be clear (and hopefully stated by legal contracts) to the
companies working in the LibreOffice ecosystem that they cannot wake
up one day and bring their development outside the LibreOffice
project. They cannot stay with one foot inside the ecosystem,
contributing to it, and with the other one bringing their development
effort outside. This is something the next board should focus on.
Given that we are doing FLOSS (Free / Libre / Open Source Software)
here, I wonder how would you like to frame such a legal contract, given
that the "right to fork" is one of the basic Free Software freedoms?  
And what if it was not a company, but another group (do you remember
IANAL so I have no clue on how this kind of contract could be settled
but I want to say that is quite logical that you cannot both support and
oppose an entity. In Italy, the law states that if you leave your job
and create a company that will do the same job you were doing as an
employee, taxes will be much higher.
From my point of view, rather than legal contracts, a much better
strategy is to listen to what the ecosystem companies or other
contributors are telling you; work with them, instead of against them;
treat them as partners, not as enemies.  If you do that, there is no
reason for anybody to leave the community, move the code away, or fork.

Clearly, this is also my view. However, listening should not mean that
the foundation does everything and only what is requested by the
companies in the ecosystem. This has to be clear.

The foundation should create contact between businesses and users, not
be the representative of a group of companies.

In the Online case, we (as Collabora) were trying extremely hard not to
have to move the development outside of TDF infrastructure, and we have
done many steps to fulfill asks of various people.  Unfortunately -
they were demanding more and more; and at some stage they wanted just
too much: to dismiss the agreement we had with TDF that LibreOffice
Online will be a source-only project, for *everyone* to build their
branded versions on, ie. it won't be a binary product that people can
download under LibreOffice name from TDF pages.

This agreement was in place to ensure that the economics work
correctly, and all ecosystem companies (not only Collabora) can build
their Online's on top of the shared code.

"all ecosystem companies" = ["Collabora", "Allotropia (formerly CIB)"]

Although I didn't have the opportunity to take part in the discussion
before the decision was made, I can say I'm fully endorsing the
termination of the agreement (which was de facto terminated by Collabora).

I used to think, and will continue to, that keeping Online difficult for
everybody to build and vendor-provided only was a huge mistake and, most
of all, as you demonstrated, did not achieve the goal of "ensure that
the economics worked correctly". People didn't have the opportunity to
"download and try" the product as a part of the LibreOffice suite and,
therefore, did not buy any support/extra subscription for it.

Getting to this specific situation, Jan 'Kendy' Holešovský, in the
last Q+A session for the next BoD, stated that “it was really hard
for us [Collabora] to get contributors and volunteers under the TDF
umbrella… and we tried hard […] now that we’re on GitHub we get
several commits from random people just because it’s on GitHub” [1].
Kendy didn’t bring any data supporting this thesis but – for the sake
of the argument – assume he’s true.
Sure - I was unprepared for the question, it was ad-hoc, so everything
I've said there was without the possibility to have the numbers at hand
in advance :-)  Let me add the details now:

d0edfeabbdc969a9a66cf90976a63c2f4403a6d3 was the last commit that
happened on the TDF infrastructure.  It covers work from 2015-03-03 to
2020-09-30.  The amount of people who have contributed during those 5
years and 7 months was 85:

$ git log d0edfeabbdc969a9a66cf90976a63c2f4403a6d3 | grep '^Author:' | sed 's/^Author: //' | sed 
's/ <.*//' | sort | uniq | wc -l

In the rest of the period, so from 2020-10-01 to now, the amount of
people who have contributed in the 1 year and 2 months was 75 (155 if I
include the translations - but that wouldn't be correct):

$  git log --invert-grep --grep=Weblate d0edfeabbdc969a9a66cf90976a63c2f4403a6d3.. | grep 
'^Author:' | sed 's/^Author: //' | sed 's/ <.*//' | sort | uniq | wc -l

Do you see the 5 years vs. 1 year proportion?

Well, yes but no. A more deeper comparison should be done here, in my
honest opinion. If the 50% of the 75 people contributing in the last
year are former developers (e.g. myself), that ain't that big success...

Shouldn’t this have been a concern for the whole foundation, and not
only for Collabora? It’s the foundation scope to bring new developers
Definitely it should be a concern - and I've stressed this several
times on other occasions that the main reason why I am standing for the
Board is that I want to make TDF more welcoming (to get new
contributors) & friendly (to keep the existing ones).  And by
contributors, I mean not only developers, but also translators,
designers, documentation authors, QA and other people actually
enhancing the project.

If GitHub can magically attract developers, also TDF, from my point
of view, should move there.
There are many pro's and con's; eg. we were criticized after the move
that GitHub is proprietary software & infrastructure.
For an open source project to move to a platform owned by Microsoft, is
definitely not the kind of choice I can say I would endorse.
Also I think gerrit works better for a C++ project rather than the
system of PR's known from Github.

But if you think even TDF should consider the option, please do feel
encouraged to research that & come up with a proposal to the ESC and
the Board!

There are indeed a few concerns I have to bring: have you ever
wondered why the Nextcloud Server project on GitHub has 1.6k open
issues? Why do they need so many tags, bots, and PEOPLE, employees
that spend their time closing useless issues that are used as support

What I’m trying to say is that a wider audience also comes with
considerable disadvantages.
I am actually not sure what should I conclude from this?  Are you
arguing that Nextcloud is doing better on GitHub than TDF on its own
infra, or that it is doing worse?

My experience with issues @ Nextcloud is definitely not idyllic... I
can't say it's only due to the fact they're on GitHub but I think this
contributes. There are many duplicates for any issue, some of them will
be closed as resolved by thinking the problem was solved when it was
not. Other are just kept there.

Can you add how many open issues are at TDF?  How many tags, bots etc.
do we have?
We could but it would not be a fair comparison since are two completely
different product. Perhaps the number of open issues per SLOC could be a
good index.
In his speech, Kendy also mentions that “we [Collabora] love
LibreOffice”. I am sure that what he says is true,
Yes, we do love LibreOffice, just have a look at the amount of commits
coming from Collabora to the LibreOffice repository :-)

which is precisely why he (with the whole Collabora team) could help
us understand why they renamed lool (LibreOffice Online) to cool
(Collabora Online) also in the source code [2], removed the “This
file is part of the LibreOffice project.” statement from the headers
of the source [3] and changed the variable names [4].
Unfortunately there is no way for me to win this argument.

If we keep those names & comments, it could be argued that we are
misusing the LibreOffice name - "why do you say that you are part of
the LibreOffice project when you are not on TDF infrastructure?"

But when we have removed it, it could be argued that we don't love
LibreOffice enough...

Luckily there are hard data that show where we stand - Collabora has
contributed 32.3% of commits that went to LibreOffice since the last
October; which is BTW on par with what we (or any other company) are
allowed in the bodies as the percentage of representatives, so it seems
to be a healthy number.

I like to read the story in another way: Collabora worked hard to create
online, put the effort of many developers per year and now finally
wanted to see only his name on it, instead of LibreOffice. But is of
course just my personal point of view.

In conclusion, I would like to emphasise the fact that I’m completely
unhappy with the “attic” proposal as a solution for the Online
situation and hope we can all work together to allow TDF to still
consider Online a part of the LibreOffice suite.
I am still not sure I understand your reasons - why it is so important
to you to name this code "LibreOffice Online"?

The code is still Free Software under the same MPL license, as it was
under TDF. 

... and has been cleared (at least to me) just on Sep 23, 2021, when
also the EULA was removed from the repo.

 The code is flourishing on GitHub, so the FLOSS world wins;
and only few people actually care if it is named "LibreOffice Online",
or "Collabora Online".

Of course, we as Collabora are actually among those few who *do* care
if it is called Collabora Online or not, because that helps us to get
customers, and re-invest the income into the code; see more here:

I know how much you care about building an ecosystem around Collabora
Online (and not LOOL), see e.g. this issue that was
magically commented by someone @ Collabora after I shared the link to it
in the legal mailing list.

Using the Collabora product name prevented and prevents the customers
from acknowledging there are other companies that provide support for
Online, damaging, again, the ecosystem, instead of strengthening it.

The last thing - have you noticed the recently announced "Nextcloud

The same Online code is now available under the "Nextcloud Office"

Please ask yourself - what is the value of non-atticized LibreOffice
Online?  What is better for FLOSS in general: One common code on GitHub
with multiple names (all promoting the LibreOffice Technology, but none
of them actually called "LibreOffice Online"), or 2 diverging forks,
one of them consuming TDF money, but called "LibreOffice Online"?

It's because I ask myself this kind of question that I spend my time
writing on the board-discuss mailing list.

It is obvious, and I thought it was also in my previous mail, that
having two divergent forks wouldn't be the way out of this problem.

Also, my main problem is also not the source code but enabling the
foundation to still include Online as a product of the LibreOffice
line-up. Do you (or anyone else) have any proposal to achieve this goal?

Hoping that someone else is willing to take his contribution to this

All the best,


To unsubscribe e-mail to:
Posting guidelines + more:
List archive:
Privacy Policy:


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.