Executive summary and analysis of the Helsinki City and OpenOffice case in 2010-2011

am: 2012-04-12

This document was prepared by the Free Software Foundation Europe (FSFE) for readers who are interested in the City of Helsinki's OpenOffice pilot project, and in lessons that may be drawn from this project.

Background: History of the OpenOffice pilot project

The initial reason for the project to emerge was a political initiative made by city council member Johanna Sumuvuori. Quoting her initiative "In the city of Lappeenranta the increasing usage of open source software¹ lead to savings of some 70 percent on the schools' IT budget." The initiative was signed by 60 other council members and presented in the city council meeting 2010-11-24.

During the summer 2011 the Helsinki city recruited a new CIO. Once he had settled he prepared a plan on how to respond to the city council initiative. In a meeting in January 2011, organized by council member Sumuvuori, to which FSFE was also invited, the new Helsinki City CIO Raitio presented his plan. First he would appoint a project manager. Then the IT department would install OpenOffice on the laptops of all city board members - a group of about 600 elected representatives who participate in the city governance via dozens of smaller boards. If succesful, the pilot would be followed by installing OpenOffice on all 21000 city workstations.

Later that winter, FSFE had some e-mail correspondence with Raitio, and in March 2011 FSFE had the pleasure to meet the project manager responsible for the OpenOffice pilot. She told FSFE representatives about her plans and about some issues they have had in trying to package OpenOffice as a MSI package suitable for automatic installation on all city workstations. FSFE also learned that most of IT in Helsinki is outsourced to the company Tieto, and that this project depended to some degree on how Tieto could support OpenOffice installations. As at FSFE we wanted to make sure the project would not fail for simple technical reasons, we offered to help with technical issues, and we also introduced the project manager personally to the officials responsible for the OpenOffice migration at the Finnish Ministry of Justice. One of these was Martti Karjalainen, the project manager at MoJ who later wrote a doctoral thesis on the subject. We also told Tieto that we could provide more contacts if needed - for example at Novell and IBM, as these companies provide commercial support and products based on OpenOffice.

Later that spring, FSFE representatives received an e-mail from the project manager Raitio, notifying us that the automated installation of OpenOffice on all the city workstations was going smoothly. After news about the pilot were published on osor.eu, Raitio told FSFE that the IT department would not issue any news about the developments to anybody and that we would have to wait for the final report to be published. When the final report eventually was released to politicians on 2011-12-28, the city's IT department did not notify FSFE.

Despite the lack of official information, in early December 2011 some news leaked out about how the pilot was going. It turned out that:

Not only was OpenOffice installed on the city boards members laptops, but also MS Office was removed, thus forcing the boards members to interact with the surrounding city administration still using Microsoft's prorietary file formats (doc, xls), with those proprietary formats causing a number of interoperability issues.
The users did not receive any training, they were supposed to adapt to the new program on their own.
Users had problems opening the newer MS Office formats like .docx and .xlsx. The board members started to teach each other to change the file endings to .doc and .xls which solved the problems. From this, it appears that support for docx and xlsx files was left out of the installation on purpose, as these files are normally supported by default.
The IT department had conducted an e-mail survey to board members asking if they wanted MS Office back, to which only 25% responded that they need it. The majority wanted to stay with OpenOffice despite the fact that there was little support available from the IT department.

Analysis of the IT department's report

A few days after Christmas 2011, the IT department distributed a report on the OpenOffice pilot to the city council. This 20-page report concluded that switching to OpenOffice would cost 74% more than continuing to use MS Office calculated on a 7-year period. According to the report the investment in a two year migration project would be 4,5 million euros, followed by an annual cost of 3,4 million euros. In total the calculation claims that migrating from MS Office to OpenOffice would cost 21,5 million euros, which clearly cannot be true. The part of the report presenting the calculation is only 3 pages, and it lacks any actual calculations, so we don't know which issues are projected to cause which costs. The city is not releasing the methods that were used for its calculations.

However, the report mentions that the calculations are based on a model presented in the document "Gartner Toolkit: The Cost to Upgrade Microsoft Office or Move to OpenOffice.org". As that report is not academic research, but a Gartner paper that costs 7500$ to obtain, it is also questionable whether the Gartner model has been a subject to proper review and evaluation since it is practically secret. All that is known about the Gartner report is that the document is written by one person alone, and the author Michael Silver gave statements favouring Microsoft's OOXML in 2008, when the rest of the industry (e.g. IBM, Novell, Google) backed the Open Document Format that later became an ISO standard and is widely supported in about every office suite and document software available today.

The secrecy alone renders the results suspect as a basis for any decision making, whether inside or outside the Helsinki IT department. While this secrecy makes it nearly impossible to have a serious debate about the calculations, a few of the assumptions presented should be questioned. For example, the calculations expect that despite migrating to OpenOffice the organisation would still continue using proprietary MS Office file formats. However, one of the biggest benefits of migrating to OpenOffice or other similar office suites is that they use the ISO standardized Open Document Format natively. Thus, the organisation can be sure that their office documents are electronically archiveable and can still be opened in 10 or even 100 years. In addition, relying on this kind of open standard brings also other benefits in interoperability, as files can be read and written anywhere, anytime, without a lock-in to one vendor. Another assumption in the calculations is that after the migration 5500 users (12% of the total computer using personnel of 30000) would still need to use MS Office, due to some "advanced features" that supposedly perpetually exist in MS Office and never come to OpenOffice or LibreOffice.

The secrecy issue is made worse by the fact, that the report issued by the Helsinki City IT Department does not mention any authors. In normal cases officials are required to have their name visible when exercising their powers. Did the authors resort to anonymity because they did not want to be held responsible for the results? Was the report even written by city officials, or by some consultant from the City's current systems vendor?

Very interesting is also that all cost analysis was done by consulting only Microsoft and their resellers. To our knowledge, no request for information (RFI) was issued to companies providing services related to OpenOffice or other office suite vendors. Neither was at least IBM nor Novell consulted directly. So the whole cost calculation lacks any price information from the vendors that would potentially provide commercial support for OpenOffice or similar office suites. Neither were any figures provided about the Microsoft Enterprise Agreement price or other costs from Microsoft. Only the cost difference was presented.

It was also interesting to note the whole notion of public procurement was absent from the report. The document's anonymous authors appear to ignore any issues related to free competition and public tendering. For example, the calculation assumed that during the whole seven year period for which projections were made, software like Microsoft Sharepoint and Exchange would remain in use. On page 7 there is a figure showing that the IT department has already made a decision to renew the Microsoft Enterprise Agreement every three years until at least 2019, which may be illegal according to Finnish procurement law.

From a taxpayers point of view, the CIO of Helsinki made a grave tactical mistake by releasing a report revealing that the city has no alternative plans at least until 2019 and that it is planning to renew its contracts with Microsoft as long as the price rises less than 3,4 million euros - all this when the next round of negotiations with Microsoft is scheduled for the 2012, the year following the release of the figure The bare minimum a CIO should do in these kind of situations, is to keep the option to migrate away on the table during the price negotiation to migitate price increases, even when the CIO would not personally like the option to actually implement alternative solutions.

FSFE will continue to follow up in this issue and support any activities that promote alternatives and competition. At FSFE, we feel that a key issue would be to properly procure software as all other tax payer funded acquisitions are. To ensure fair competition and efficient use of public funds, no IT department should take Microsoft products for granted, nor OpenOffice either. Each system should be procured and the best offering chosen.

If you feel we are working for a good cause, please help us by:

submit to us copies of any Microsoft agreements your local governance has signed
sign up as our supporter at fsfe.org/support
secure our funding and donate to the FSFE