Keeping a promise made a long time ago

Some time around 2009 or 2010, the OpenDocument community realized that while it had won the moral battle over Microsoft and its dubious OOXML standard, it had lost the adoption and ecosystems war.
Microsoft Office had been released and with it an undocument format called OOXML which, as far as experts were concerned, had little to do with the ISO 29500 (aka OOXML) standard. While Europe and Brazil were struggling to migrate their public sector’s documents to ODF, any company or government, let alone any individual acquiring Microsoft Office 2010 migrated to the new and shiny OOXML, officially without remorse or complaint. The ODF advocacy groups here and there were launching all sorts of events and meetings to guide and assist migrations to ODF. Results were mixed. We had victories. We had defeats. At the end of the day what was at stake was fear of failure and change from CIOs and IT services. That’s still the case today. But while these are mostly human factors, there is one thing we hadn’t tried yet, or at least hadn’t been tried enough: turning the hundreds of thousands of files that are out there and locked up in various proprietary file formats to ODF documents.Docliberation

This week the Document Foundation announced its second major project, the Document Liberation. Its aims is to pool and collect every file format filters we have and that people are willing to contribute and develop them so that they not only keep improving but are distributed in the largest number of applications. The aim of the Document Liberation is thus simple: to enable everyone to own its content and to bring a solution to vendor lock-in and undocumented file formats. In doing so, the project is keeping a promise made a long time ago, specifially by ODF. But ODF is a format itself, and while it is enjoying a pretty widespread adoption, it has not done what Microsoft did with OOXML: propose a smooth transition through a change people can accept. In the case of OOXML, as lousy as it seems, people accepted the change because they didn’t know better: Microsoft does this for a reason, so things will pan out all right in the end. They’re taking care of my documents. The industry will follow.

In the case of ODF, no one was in such a position, except perhaps Microsoft. The approach we’re taking today is to offer a solution to a very real problem millions of people have: they don’t know what to do with their files if they haven’t migrated them to a more modern, but not necessarily more standard or more open file format. To these users, we offer a range of choices depending on their “predicament”. We will add more filters as time goes by and the community grows. To developers we offer an exciting place to contribute code by improving existing format filters and proposing new ones. To everyone we offer code that will ensure the continuity of access to content locked in countless files scattered across the Internet, personal computers, corporate, academic and governmental archives. In doing so we not only help ODF keep its promise to liberate documents once and for all, we help make the world a better place by empowering everyone to access and create more digital knowledge and unleash creativity. This promise lies at the core of the Document Foundation’s mission.

Document Freedom Matters

As the Document Freedom Day is approaching I realized that we don’t push ODF and open standards as loudly as before. Certainly most of the battles for the mind and market share are past, at least when it comes to office file formats. But the recent public consultation of the UK government brought back some of the most crucial issues surrounding ODF and it’s useful, I think, to check where stand these days on these matters. dove-logo-2012

Shortly after OOXML had been given the ISO label in the weirdrest and most outrageous way, a representative from Microsoft spoke at an IT conference in Brussels and bluntly declared “ODF has won”. Well, it is true in the sense that the OOXML standardization process had highlighted the -probably terminal- inability of the ISO to tackle IT standards and transparency in an effective way. In this sense indeed, ODF had won the “moral” battle. On the other hand, Microsoft had reached its main goal: to get the ISO’s stamp of approval on the half-baked OOXML specification. What it never achieved however, was the actual mass adoption of OOXML by the users of Microsoft Office and beyond. What’s that you say? OOXML has not reached mass adoption? Well, the actual specification, the standard, has not actually been implemented as the default file format until MS Office 2013, and yet, as an odd, end of the list option called “OOXML – strict”; OOXML – Transitional being a rubberstamp for every undocumented and unpublished sub-spec and binary blob necessary for Microsoft Office’s secret sauce to work its magic.

Of course, it is sad, yet true, to notitce that the mass adoption of ODF has failed until now, and that most companies and government use Microsoft file formats (xml based or not) as their default format. This alone would be enough to claim we haven’t moved an inch closer to true document freedom. At the same time, the IT and the way we use software and data has changed in 5-6 years.

More and more data
Data, big and small has become the new black. It’s all about data these days, and yet amidst all of it, the world does not seem to care a lot about digital documents. Any true reality check would have anyone notice that office suites are used everywhere, not less, not more, but in addition to any online collaborative, social and multi-device service and platform outt here. But most of this data is not wrapped in an office document format though: the office document is an aging, yet useful metaphor that is currently not explored too much. In this sense, the mission statement of OOXML goes against the wind: a standard to represent and encapsulate the information and its representation contained in proprietary documents is not a solution. It’s one more layer of complexity: better open up all the specs of the former proprietary file formats instead!

No growth in the ODF ecosystem
Like it or not, there has not been that much investment in the ODF ecosystem. Of course, work on the further development of the ODF specification has continued at the OASIS Consortium. But that alone is not enough. Once you have about 4 office suites implementing ODF in an good or excellent fashion, you need tools, and you need middleware. I know a few tools, some of which my former company, Ars Aperta, has contributed to, but you also need middleware such as content management systems, specific application servers, data busses, etc. to implement and support ODF. Beyond that, you need good mobile support for ODF. To this day, this has not happened yet. Although I know of serious attempts, this is just not available and won’t be, I guess, until the Document Foundation releases its own ODF viewever for Android and iOS, which is thankfully actively developed. Once you have all this, entities migrating to suites like LibreOffice need to have an actual document exchange policy that takes into account exchanges inside the organization and outside of it. Most of the migrations I know of do not integrate this and as a result I have heard of successful migrations to LibreOffice with no or only partial document migration to ODF.   In a nutshell, ODF is no icebreaker. It is an asset, it is an opportunity and a great standard for office file formats, keeping your data safe, accessible for edition, consultation and archival. This alone is a major advantage. But what was seen as the strategic weapon against Microsoft dominance on the market seems to have failed.

Reasons to hope and fight for
In all this rather grim description, one can wonder why we even have a Document Freedom Day at all. It is perhaps important to realize, first of all, that this is not a grim picture. The very fact that the ODF ecosystem stands the way it does today and has survived until today in the face of aversity and corporate friction (monopolistic practices, poor procurement policies, Sun’s acquisition by Oracle, etc.) is the living proof of the ODF’s ecosystem resilience and strength. Second, ODF is being adopted in many places around the globe, and it works. The Brazilian Government, entire parts of the German public sector including the famous migration of the City of Munich, the French Gendarmerie, the Portuguese and Italian Governments and other public sector agencies all around Europe, as well as small and large companies here and there do use ODF. You can send letters in ODF to most of the European public sector including the European Institutions and they will be read and handled just like .doc or .docx formats. What this means is that while ODF may not have won over proprietary formats we now have a growing place and use for opens standards like ODF. And as the current IT trends towards the reuse and digitization of cultural and public data go, the open data movement as well as major projects such as the Preforma FP7 project are true opportunities for growth, change, software and document freedom.
Last but not least, the advent of the LibreOffice project is probably one of the best news for the advancement of ODF. Because it was succesful in getting rid of the corporate overlords dominating the old project, LibreOffice managed to set up an increadibly fast growing project and community that managed to develop innovative features that ultimately benefited ODF and software freedom as a whole. But we cannot stop there.

Software Freedom matters, today perhaps more than ever. We need to regain control of our software, of our data, where it goes, what it says about us, whether we will be able to use this data in 30 years from now  and what we can do with it today. We need Software Freedom and we need Document Freedom. Let’s fight for them.

Why LibreOffice 4.2 matters more than you think

On Thursday the Document Foundation released its newest stable branch, LibreOffice 4,2. Don’t let be misled by its number; if we weren’t on a strict time released scheduled alongside a clear number scheme without any nickname for each release, I would have called this one the 5,0. Yes, you read that right, the mighty Five. Why? Mostly for two big reasons.

This is a major code overhaul

Do you remember one of my first posts about LibreOffice, at the end of 2010? I had hinted that one of our goals was to develop a brand new engine for Calc, which had stayed pretty much the same since 1998. Well, the 4,2 just got that: Ixion has been integrated as the Calc engine and that, among other things, such as real-time integration of data feeds, is about to change a lot of things, and not just in terms of performance boosts (over 30% of improvement depending on the cases). This might actually open the door for brand new types of users in professional and scientific venues for instance.

Alongside this rewrite, we also have a major work on the user interface layout and dialog rewrite. As Michael Meeks explains it, we had introduced this rewrite with the 4.0 but now quite many of our dialogs and widgets have been rewritten. Other user interface improvements such as a brand new iconset, document snapshots on Windows bring a fresh and refined user experience to LibreOffice.

More Enterprise-ready than many others

We have heard this song here and there. You cannot be innovative and be successful as an enterprise solution. You cannot be the right choice for companies if you haven’t a major American corporation as your main sponsor/steward/overlord/friend. You cannot deliver a professional grade office suite if you work along a time-based release system. I think that these theories have already been proven wrong, unless you have a twisted definition of what the enterprise market  needs. But with the 4.2, we also have some nice and immediately actionnable features that will appeal specifically to the enterprise market:

  • Integration of the CMIS stack allowing you connect to document repositories on SharePoint, Nuxeo, Tibco, Alfresco, Google Drive and many other CMS.
  • Expert configuration options now all put in one place
  • Advanced deployment options
  • Better group policy controls for deployment and installed user base
  • Improved Microsoft Office (.docx, etc.) and RTF document filters
  • Improved look and feel on Windows
  • Change tracking on ODF and even on OOXML documents

I’m not listing a good dozen of other improvements of importance, but here’s the complete list.

It’s not about success. It’s about what comes next.

And now, I’m going to really explain why LibreOffice 4,2 matters more than what meets the eye. The amount of code clean-up, refactoring, write up, the inclusion of new features and the continued growth of contributors between the moment the Document Foundation released LibreOffice 4.0 and the 4.2 is truly amazing. The 4.0 was a major accomplishment, but this time we did even more, seemingly with less effort (although this comment does not diminishes everyone’s accomplishments for this release).

What’s going here? A giant in Free and Open Source projects is emerging and we are witnessing this unfolding right under our eyes: a growing development powerhouse, increased funding, an effective structure, overworked but growing contributors, an increased presence on worldwide events, improved processes on localization and quality assurance… I guess many observers as well as several insiders were thinking that once we had set up the Document Foundation as a structure and released the 4.0, things would take a course and a pace of their own. That hasn’t happened. On the contrary the word around the project was “Up!” and has not changed ever since. Another  possible reason is that once the founders -with some hindsight, I start to see it more clearly- got the structure, the governance, the main processes going, priorities started to change for the best: Discussions started to be more about resources, funding, sustainability, but the minds were freed from the worry of the next day and were able to focus on developing something great. I realize I’m painting a very nice picture, but I know that the road won’t be short and it will not be easy eiither, but judging by what this community has already overcome I am confident the Document Foundation is going to push the enveloppe on many levels in the years to come. I am truly proud of what we have accomplished so far and I would like to thank everyone who made this release possible. Happy FOSDEM!

Forget about meeting customers’ expectations: Innovation comes first

… and so does pesky market research. The IT bubble has been spreading the word about this Forrester report and as you can imagine it got many of us wondering what it really means. Well it got me wondered about a few things too, but perhaps not for the same reasons others twisted their heads around. Let’s start with a good one.

“OpenOffice derivatives ” at around 13%?

I mean… really? It’s not that I don’t believe the number – I’d stick it around 15 to 17% actually but… so be it. The funny thing here is that since 2001 (the birth of no such study had placed any “OpenOffice derivative” at that number. We heard 2%. We heard 5%. We never heard of anything like a double digit. And even more incredibly, that number, 13%, went down to a combined total of 5% of market share. Unless no one of us at the Document Foundation and perhaps at Apache caught that in 2011, nobody knew Forrester had seen that market share. So it all looks like we learned about these good news 3 years later. The other funny thing is that until this study, many analysts in the industry would have told you that such a double digit number (and granted, not a really big one) would have gotten Microsoft shivering and the whole IT press circuit on orbit for 3 months just with that percentage. But for some odd reasons it went completely unnoticed.

“Google Drive: 13%”

This is the big news and here again I’m not surprised by the number. But where’s the catch? The catch is a double one: most organization I know of use Google Docs in conjunction with another office suite, and not a cloud-based one, mainly for collaborative work inside the company. It’s thus important to realize that kind of scenario is out there and is far from being a minority one. Granted,  there report mentions that multiple answers were possible. The second very important point is that if customers have been switching to Google Docs in this proportion, then it confirms what many office suite migration experts will tell you (privately): most of the migration is about change management. Forget about that “very important VB macro the accounting and Joe from Sales can’t do without”. Overnight the new kid on the block has changed, it’s called Google Drive and it sits on a cloud. The macros and the document fidelity all seem not that important in hindsight (just try several formats in Google Drive just to see what I mean by the relative lack of importance of document fidelity).

In other words: stop listening to the customers and innovate.

Oh yes, and it’s not just because Henry Ford once said it. What about this large administration that refused to migrate over to LibreOffice because they felt the pivot table support in Calc was just too different, if not as good as the one in Excel? Well, they’re on Google Drive now, and the only thing left to do is to wish them good luck with testing the comparative strength of each pivot table support.

The real challenge is for LibreOffice to expand to the cloud and to the mobile platforms while at the same time find the right solution to let users use whatever storage they want: various cloud services, a mobile storage provider, or even a peer to peer or unhosted technology. To be fair, this report is by no means a wake-up call: we know the need and how critical it is. However it  is somewhat of a useful whip for all of us… and an invitation to grow our resources to do more and reach further for the benefit of all!

The Spin-Off

The news are out and while they do not concern the LibreOffice project directly (insofar as this does not represent a change within the Document Foundation) I thought I’d be explaining a bit what these news are about and what they mean for the LibreOffice project. First, the announcement, as well as Michael Meeks’ blog post on the subject. (Caveat Emptor: I dont’t work for Suse and this is a personal blog, expressing my views only, although I’ m obviously a member of the Document Foundation’s board of directors which makes me anyway speak from a different point of view from the guy next door) Essentially, the story can be summarized in this sentence: Suse wants to focus on its core business (server and cloud computing) and has decided to spin off part of its LibreOffice development team over to Collabora.

What this means is that the Document Foundation has won a new strategic partner, Collabora, and that it has now a seat at the Advisory Board. What this also means is that Collabora has joined the ecosystem of LibreOffice service providers and that it will de facto stand as one of the very best companies making money on LibreOffice. Of course this does not automatically ensure their business success, this is why we wish them all the best.

Some readers might be surprised by the way I’m describing the announcement. It might be tempting to see these news as the sign of the upcoming demise of the LibreOffice project. This is very far from being the case  and there are two good reasons for that.

First, Suse is NOT dumping LibreOffice developers away. The same developers basically went in a new company and there working on LibreOffice development there. In American parlance, this is called a spin-off. Second, and perhaps more importantly, the Document Foundation has precisely been created to prevent not  just a take-over from one company of the entire project, but actually to ensure that any company or any developer leaving the LibreOffice project would not jeopardize the project, the community and their independence. Just as an example, today the Document Foundation has its own release engineering team and has its own infrastructure. We do  not rely on third party sponsors or on the benevolent will of one company for any of this. This simply means that if the sky falls tomorrow, we will still be up and running no matter what change in affiliations could happen then.

If anything, what has been announced today by Suse and Collabora is the perfect illustration of why the LibreOffice project is needed and why the Document Foundation has been created. I am thus thrilled to welcome Collabora to the Advisory Board and I would like to send my deep and sincere thanks to Suse for their past and continuing support, and last but not least, wish good luck to Michael Meeks and his team for their new roles at Collabora. The journey is going to be exciting.