Some propositions for TDF's future

Hi all,

In my opinion, resolving the current crisis will require elections and changes to the foundation’s bylaws and operations:

1/ The current system of BoD’s election organizes competition between directors. People are chosen by majority preference, which doesn’t guarantee that they have the slightest desire to work together. In my opinion, this system should be replaced by a list ballot, with a clearly identified head of list. That would have the advantage of forcing candidates to agree, before the election, on a project and a way of working together.
The difficulty with a list ballot is that it’s better when there are several competing lists. So you need a lot of people. Having said that, we currently have the same problem: we need 10 elected representatives (7 titular members + 3 substitutes) and there are rarely more than 10 candidates…

2/ We need to deal radically with the problem of conflicts of interest: a company that has an employee at the BoD should not be allowed to respond to a call for tenders from TDF. To prevent circumvention of this constraint (a company subcontracts a contract to a company represented on the BoD), the Executive Director should be allowed to eliminate a bidder if he discovers or suspects this kind of arrangement. This would force companies to show that they’re doing the job themselves, which seems to me to be a good thing for expanding the LibreOffice ecosystem.
In my opinion, simply forbidding a director to participate in the awarding and follow-up of a call for tenders (and then in the follow-up of the contract) to which his company responds is highly insufficient, because it doesn’t prevent pressure and negotiations and exchanges of goodwill.

3/ Contract award management lacks transparency: if we consult the BoD’s award decisions, we see that they only mention the number of bidders and the name of the winner. Most of the time, we see that 2 responses were received and that the winner was Allotropia or Collabora. How can we not suspect that when Allotropia wins, the other is Collabora, and conversely when Collabora wins, the other is Allotropia? What’s more, how can we not suspect that there’s some sort of agreement between 2 companies to share the TDF market?
In my opinion, the list of companies bidding should be systematically published for each call for tenders.

4/ We need to protect the LibreOffice team from any pressure they might come under, especially the Executive Director if his power of veto over contract awards were implemented. So I think that the dismissal of the ED and the sacking of a team member should only be decided by the BoD after the assent of the Membership Committee.

5/ last but not least, I think that parity should be imposed on the BoD and the MC. Given the (very?) low proportion of women in TDF, this will take time, but I really think it should be a permanent objective for TDF.

Best regards.
JBF

4 Likes

HI @jbfaure,

So yeah, “list ballots” will not be realistic at all. A more moderate and workable fix would be to switch elections from STV to cordocet voting (like KDE does). But even that requires changing statutes, which is a huge pain.

Which problem of conflict of interest? Has there even been one case where a tender was award to a bidder for a price that was higher than what TDF would have been able to buy on the market? I dont think so. A lot of people who should have better things to do discuss how many angels fit on the head of a pin here.

Removing the last companies willing to bid on TDFs tenders from being considered … is not a great way to “expand the LibreOffice ecosystem”. Not only will it frustrate existing ecosytem companies, it will also be a big red flag for any other company that might have considered to work on LibreOffice.

I share the goal, but how do you envision that to be done in practice? Force half the trustees that identify as female to stand as candidates for the board? And the other half to stand as candidates for the MC?
That goal is great, but not actionable in itself. Do you have an actionable proposal?(*)

Best Regards, Bjoern

(*) FWIW, I think that is something that might be more of an objective or key result for the board (or parts of the board) to aim for. As of now, it is not something very suitable for putting into rules or procedures.

Rather than focusing on the details of your statute changes, I’ll make some meta-comments or side-comments…

The current situation in the BoD and perhaps the BoT is such, that a declaration such as yours may be interpreted as a belligerent attempt at usurpation, i.e. it might be read as “I want to depose Thorsten, Gabor, Cor, etc.” - even if the personal is not what motivates your post. While I may be naive as a new(ish) member, I believe we should strive to establish some basis of collective informal agreement before proceeding to put forward changes to the statutes.

More specifically, the question of the concerns Allotropia, Collabora and/or other companies have about tender budget allocation should be addressed as a precondition - in the sense of reaching an informal trust regarding how the TDF will normatively handle these matters even without company employees/officers at the help of the TDF BoD. i.e. that it won’t be a situation of “us” non-company-employees trying to screw the companies over somehow through this policy. If we could reach that, then I assume that stronger formal CoI avoidance, election mechanism changes etc. would not seem like a declaration of war and could conceivably be adopted.

There’s also the matter of the composition of the BoT. The LO ecosystem commercial companies naturally employ a large fraction of the LO contributors, and thus also a large fraction of the trustees. And these trustees have emotional attachment / loyalty to their companies to some extent; and are more likely to vote . This, while other contributors for whom it is not the “day” job are less likely to even be trustees. I don’t know what the ratio of ecosystem company employees/subcontractors to other contributors among the trustees, but assuming there are a lot of the former, you also need to take into consideration that your suggestion may be construed as “weakening” them within the TDF. Now, I don’t think that really weakens them, but it’s a matter of perception of the situation, of the nature of the TDF, of the intents/agendas etc.


Finally, about male/female parity and other issues of representation (e.g. Europeans vs non-Europeans perhaps?) - this is certainly a problem. But it’s much deeper than could be addressed by establishing a quota. Look at our board of trustees: We have, what, 5% women? That’s dismal; and unlike in a large population that has a large fraction of women, you can’t just use affirmative preference to rectify the representation. You would essentially be forcing all female trustees to serve on the BoD or the MC :frowning:

So, this is another serious issue which I believe can’t start with a statute change but with something else. And here I don’t have any concrete suggestions, although perhaps I might focus on regional/local user groups in terms of getting more women into the “funnel” which eventually produces some trustees. And of course, input from women trustees on this matter is of crucial importance.

Hi Jean-Baptiste,

Great to hear from you. I’ve been away on a complete dis-connected
break for once (lovely) and was surprised to see quite so much mail on
my return. So forgive me if I miss the context.

2/ We need to deal radically with the problem of conflicts of interest:
a company that has an employee at the BoD should not be allowed to
respond to a call for tenders from TDF.

This issue seems to me to be routinely inflated and used to tar
directors with suspicions of suspicions and to extend the area they of
things they’re excluded from unreasonable in my view. So let me share a
chunk of reply publicly that I somehow sent to an internal only list
while replying to an older public criticism from Andreas Mantke on a
different topic - perhaps it helps to have a few facts =)

Board members associated with companies who were bidding on TDF tenders
have been completely and successfully excluded from any details[1] of
competing tenders, the selection process, who voted what way, what staff
said, what legal counsel was provided, they had no insight into
contracts and decisions made. They were completely recused.

That is/was appropriate and right, our statutes and common sense made
this clear. Obviously we can make the rules there even clearer and more
helpful too - why not?

However even better than the written rules - there was no leakage from
directors or staff involved that I ever saw: no quid-pro quos, nothing -

However - there are lots of details that are not accurate floating
about around this old process. I have only part of the picture but here
is my response to some of the mis-presentations I’ve seen. I think we
need to avoid some massive, post-truth set of accusations of various
(often mutually contradicting) bad things, let me take a few of them;
lets take this one:

On the other hand, the commercial company has to assume there are
other competing companies in the process, so every bid is (and has
to be) risky and as cheap as possible.

Please stop kidding: I currently know only two commercial companies
that are able to bid on LibreOffice source code tender. Thus there is
no competing market yet. It’s more of a monopoly / oligopoly market.
And nearly everyone knows about the formation of prices in such market
from her / his daily experience.

This is factually wrong - see list of people / companies below. But
more simply - many people from their daily experience know to get two or
three quotes when they buy a project to get a fair market price rather
than just going with the first number they see. Usually you don’t need
ten quotes.

Let me put some rough, perhaps not already public, numbers to this -
C’bra has bid on ~24 tenders and won ~7 - ie. under 1/3rd of them. We’ve
been in business for ~10 years. A rough adding up of what TDF has paid
in total comes to some small (single digit?) %age of both TDF’s income,
and Collabora’s. That seems neither excessive nor under-contested to me.

As for loosing money on tenders it is worth reading:
Winner's curse - Wikipedia - it is easy to see
tendering as a one-bid auction for the work. I wonder what experience of
bidding for tenders, and estimating the associated risks you have.
Signing a fixed-price contract to deliver something like this requires a
really deep understanding of the problem area to manage the risk.

One of the interesting things here though is that while -normally- you
might expect there to be bad feeling & concern between parties winning
and loosing tenders - indeed I hear of people litigating these things
left and right - at TDF we have something like the opposite.

I’m happy when TDF actually manages to execute and spend some of its
(growing) cash-hoard to improve LibreOffice. I’m grateful to the
tendering committee for getting on with it. If the work is won by
someone else, selected in a fair process - all strength to them.
Hopefully they didn’t get the winners curse too badly and it doesn’t
negatively impact their business. Hopefully we get the full
functionality vs. exactly what is tendered (also a problem).

At TDF - interestingly - what hostility and aggravation there is on
display appears to be directed against the suppliers even though they
deliver great work, often adding extra unpaid work to make it better
than what was asked for, showing flexibility and so on. As I think
Laszlo captured well: having passionate and competent people who love
the project and have similar goals around this problem space is a great
strength for TDF.

Some of the concerns articulated here sound like catch-22’s when
juxtaposed:

  • we want a more diverse ecosystem and project

  • companies should not bid on tenders if they have affiliated
    staff on the board (despite the very effective separation of
    affiliated staff from awarding tenders)

  • we need more people / companies to submit tender bids
    (which cost serious money and risk to create)

  • we also want to reduce the budget for tendering

  • we think we’re paying over-market rates

  • but somehow bids are also too cheap for others to compete,
    so we need to incentivize new people by paying more.

And so on, and on it goes for years.

TDF needs to decide what goals it wants to achieve over a certain
period, and then adopt a positive and effective approach to delivering
that. It also needs to properly include people in its community and
governance in line with its statutes wherever possible.

TDF should aim to forge a healthy and viable relationship based on
a fair exchange of value with all of its suppliers. Burning the trust
that makes those relationships constructive is deeply damaging.

Beyond that - TDF has had a lot of contracted suppliers, and
respondents to tenders; that’s public but hard to dig out I think we
have had at least: around half of the staff - plus Lanedo, Igalia, Khaled Hosney,
Laszlo Nemeth, Markus Mohrhard, Armin Le Grand, Collabora,
CIB, Allotropia, Marco Marinello, Hypra, and more I forgot.

As regards advertising the tenders widely - I’m sure TDF could
execute on that better - I believe TDF already incentivizes new
suppliers to try to get them skilled up and to gain diversity. Given the
size of the available budgets - I imagine this is a challenging ask.

Hopefully there are some useful facts in there for the community to
think over, and perhaps get a more balanced view here.

There are much more serious problems at TDF than tendering.

Regards,

Michael.

[1] So IIRC there was one single embarassing lapse of this - which one
board member was responsible for - sending a single +1 style vote E-mail
from you on a tender to the whole directors list: I have no idea how it
happened; but in 10 years that’s a good record of isolation.

Hi Sweetshark,

Condorcet method will not solve the problem of directors who don’t really want to work together. It is useful to choose one person, not to build a team.

Yes: for a list of 10 persons, 2 women in the next BoD, 3 in the next, 4 in the next and 5 in all the next.
The quota with list ballot will oblige all contenders for the BoD to seek to bring women into the foundation, and to establish welcoming procedures and a welcoming climate, instead of waiting quietly for decades for society to evolve in this direction.

1 Like

Hi @jbfaure ,

STV strengthens factionists that are only supported by a minority over those that are moderately supported across the board. Thus it will give you a board of warring factions if it can. Condorcet doesnt do that. Ultimately, at this point though any change to voting is a technical solution to a social problem and wont solve the core issue.

If we had 20 candidates including 2 women standing for the board (which is the minimum for a “list voting” to make any sense at all), I think most of our leadership troubles would be gone even without tweaking the statutes. That has never happened in the history of TDF though. And the losing list would not go away, but just be 10 people making the boards life hell.

If you want to make the life of directors hell, being on the board helps. However, it has been demonstrated to work just fine from outside the board too.

Best Regards, Bjoern

Picking only one important question now:
What would you suggest to do with a director affiliated to a non contributing company, that is having interest in TDF providing a product, that is competing with TDF’s own ecosystem?

It would be useful to provide a bit more information about that “important question” as the product in question is LibreOffice OnLine and the director would be myself.

We know that LibreOffice OnLine ceased to exist nearly 3 years ago when the ecosystem company decided to fork it and to bring it back to life only a decision by that ecosystem company to provide a community version would make it happen.

We all know that various actions to avoid bringing back LibreOffice OnLine have been taken by directors affiliated and associated with that ecosystem company in full conflict of interest.

We all know that last year directors affiliated and associated with that ecosystem company wanted to find me guilty of having a personal interest in LibreOffice OnLine even if it didn’t exist since 2 years. That’s well documented in tdf-internal.

So the other interesting question would be:

What would you think if you found out that directors affiliated and associated with the ecosystem companies:

  • influenced the process and voted to kill LibreOffice OnLine
  • influenced the process and voted to kill a developers proposal that didn’t include a clause that stopped TDF’s developers from working on LibreOffice OnLine
  • influenced and voted on a new developers proposal that circumvents legal advice and locked developers in certain areas so that they would not work on LibreOffice OnLine
  • influenced and voted on an evaluation of personal interest on LibreOffice OnLine which was killed 2 years earlier
  • influenced and will probably vote soon on a process of finding a permanent CoI on LibreOffice OnLine, killed 3 years ago, to lock out a director from dealing with employees and ecosystem matters and denying him a public hearing to evaluate the matter in a transparent way

Now that there is a bit more information about the “important question” it might be easier to evaluate a series of answers.

I think that LibreOffice OnLine, as it was, is dead and unless the ecosystem company wants to bring it back in some forms, TDF should look at other technologies and protocols to allow for collaborative editing with the desktop version and possibly other ways to have it running in a browser.

That could still create issues at that is a market of interest for some ecosystem companies but this time hopefully we could draw clear agreements that allows TDF to support projects to which the community contributes and remain available to the community.

Ciao

Paolo

Correct. And you tried to push TDF in the direction to compete with the ecosystem, with no good reason then I think your own business. Given that situation, it doesn’t have to surprise anyone that Collabora choose to move the development of online to a place safe for that sort of threats.
You may want to educate people on the success you had with your visionary plans conquering markets with your former partners Kopano and Nextcloud?

You know that is nonsense. And none of the activities to do anything on developing online under TDF’s flag has been counteracted. IIRC, the first patch of the team enthusiastically pushing the development of online in TDF with you, has even been made suitable for committing by one of the professional developers.
It is all really too sad.

Hi @PaoloVecchi, @cornouws ,

I think at the core of LibreOffice Online not becoming a success is that TDF failed to build a sustainable, independent and meritocratic community for the international development of free and open source software based on open standards around it. And it didnt succeed in doing so because that is not trivial, not because of “ecosystem companies” sabotaging it.

TDF never seriously or successfully lobbied or marketed for contributors there – neither for commercial entities nor for volunteers contributing to the effort.

In effect, it seems some people just wanted a LibreOffice Online product without bothering for creating a collaborative effort (aka a project). And thus the ending was inevitable.

Anything fundamentally flawed with that analysis?

Best Regards,
Bjoern

I suggest you upgrade your definition of TDF’s ecosystem.

We can use whatever (placeholder :wink: ) term we want, Jean-Baptiste.
But as director I cannot imagine that either trustees nor users making donations to the foundation, would be happy with the foundation spending money to compete with the companies that contribute massively to the vital development of the software.
And I do not remember that I’ve heard any of the founders making statements or suggestions in that direction either.

I’m not up to speed regarding the whole LibreOffice Online topic. Anyhow, LibreOffice as local installation is losing relevance at Enterprise level. This not good, if you expect the Enterprise level as source for new contributions

However, I must note: I’m still having hard time envisioning of code contribution by Enterprises by themselves. In the sense of having a developer in-house. Especially if it’s not you’re core business. You likely hire someone who is familiar with the code and on ‘project’ basis.

LibreOffice will lose even more relevance/market share, without Online. Access to all the files on any device at any time is becoming the standard. However entering this market is tricky. Currently you simply ship software package. Online entails lots of services. Cloud storage/ maintaining, data protection.

Entering means also operating in the market where an eco-system partner being active. The TDF product will compete with the one of the eco-system partner. if it’s same of similar product. It’s might cause redistribution of Enterprises users to TDF, without adding new users. So cake stays the same. No improvement below the line, IMHO. So dammed if you do, dammed if you don’t

TDF could theoretically also re-sell the online version with commissions (a lot less effort). However everything would become even more interwoven (entangled), likely with an (specific) eco-system partner.

OTOH, LibreOffice can’t do without the current large contributors. TDF members see TDF as self-reliant, neutral, independent party with it’s own agenda. At the same heavily relying on the (large) eco-system partners (in my perception: Collabora/ Allotropia). I’m quite often asking myself the question: what’s would be left, if the would seize contributions? A major part of the core would be gone: people who do the coding (and do code reviews). I foresee a quick decay. I think the eco-systems partners would struggle to survive too - in the long term - without the contributions from the TDF community, though.

1 Like

Hi,

note that Collabora and Allotropia were not the only companies who sponsored upstream development: All of e.g. Novell/SuSE, Canonical, RedHat did so too – and slow and steady these commitments got reduced.

Best, Bjoern

Hi @Sweetshark,

We all know that it is not trivial and I never said that ecosystem companies wanted to sabotage LibreOffice Online… until the fork., after is a very different and very well document story.

I’m not sure what you did while you were a director to promote and support what the ecosystem companies were doing but I’m the one that pushed to start the marketing and communication plan that gave and is still giving a lot of visibility to the ecosystem companies.

It is sad to see that people still believe what you stated but I guess it’s more convenient to accept an easy to digest narrative instead of looking at what really happen. I think you should look at your BoD minutes, if you still have them, to remind yourself how TDF helped that project being known out there and what actually stopped other people and organisations contributing to it.

So, based on the evaluation of documents, emails and minutes your analysis is fundamentally flawed.
I’m sure you will come to the same conclusion if you recheck your archives.

Ciao

Paolo

Hi @PaoloVecchi ,

I think the board tendered the kick-off for online for a start, laying the groundwork for the technology – without that LibreOffice online wouldnt even have existed. Note that while there were hot discussions back then too, the board worked together whenever possible in those days and not against each other. But yes, LibreOffice didnt have a good strategy for online forward from there that we could commit on back then (and obviously there still isnt one).

If you claim that as an individual effort and not one of the board as a whole, allow me to check back with the ecosystem companies on your performance on this one – after all they should know best about the results.

Best Regards, Bjoern

Hi @Sweetshark,

That is correct and promoting it as LibreOffice OnLine gave the project the visibility needed to build a viable commercial proposal for the ecosystem. So it seems we all agree that without TDF an on-line version would not have started and would not have become a successful proposition for the ecosystem company.

I read the minutes of the time and yes it was agreed that it was important to support the growth of LibreOffice OnLine but what was missing was a clear agreement about the way forward once the project became commercially viable. It seems like 3 years ago it reached that position and the LibreOffice OnLine brand was not needed anymore.

I stated that I was the one pushing for the start of the marketing and communication campaign which was supposed to further support the ecosystem companies while also clarifying the relationship in regards to LibreOffice OnLine.

I even got some resistance from a director affiliated with the ecosystem initially but then it has been accepted by the whole board, the project started and then it has been handled by our marketing team.
I did not get involved with it since then.

Ciao

Paolo

Hi @PaoloVecchi,

Yes and no – such an agreement could not be created because the prerequisites where not there. What was missing was a clear interest of the LibreOffice community to contribute to that effort in a relevant way. All expertise and all work had been done by a single vendor – way beyond what that meager tender would have covered. The community did not add any expertise, contributors or sustainable business models elsewhere.

If that would have happened, it could have served as a base to build a project around. It would have been up to those interested in using LibreOffice Online to foster more contributors for it. That never happened.

Best Regards, Bjoern

Hi @Sweetshark,

I’ve been told by several sources that many people did not contribute as they had the feeling they would contribute to a third party product, although the company was a company of the ecosystem, instead of contributing to a community effort.

When I tried to clarify the situation and see if we could actually rally a community around it, while also supporting the ecosystem by pushing to start the new marketing plan, we got an answer which supported the feeling of those that avoided contributing.

Anyway I was not in the board when LibreOffice OnLine grew enough to become commercially sustainable, thanks also to TDF and the community, so I’m not sure what you or the directors affiliated with the members of the ecosystem could have done to improve things and avoid actions that then led to a fork.

Ciao

Paolo