Also available as a PDF
The FHISO Board of Directors would like to take this opportunity to update you, our members and other interested parties, on our activities over the last year, and our plans, as well as our concerns, for the future. This is the first time we have prepared a report of this nature, covering all aspects of the organisation, but we plan to do this annually in the future.
This report is arranged in three sections: the first section covers our technical progress drafting genealogical standards and explains how they fit into our overall vision; the second section discusses the organisation’s governance structure and announces our intention to hold board elections this year; and the third section summarises FHISO’s finances and explains our decision to start collecting membership renewal fees.
Technical
FHISO’s mission is to develop open standards on the digital representation, processing and exchange of data by genealogists and family historians worldwide. It is not our intention to create a single standard encompassing all of genealogy: no-one has succeeded in doing this since GEDCOM in mid-1980s, and expectations have grown considerably since then. Instead, our strategy is to create a series of component standards, each dealing with specific area of genealogy in depth. Our first component standard will be an extensible framework and vocabulary for encoding all the elements normally found in citations of genealogical sources. This is a significant gap in existing technologies, and informal discussion with vendors suggests there is a real desire for further standardisation here.
Component standards need to be usable with a variety of technologies, and we have identified GEDCOM and GEDCOM X as the two of most important host languages. GEDCOM X is inherently extensible, so will readily accommodate data from our component standards, but GEDCOM is not satisfactorily extensible; it also has not been maintained in the last 20 years and is out of date. We therefore plan to produce a serialisation format and data model which is fully compatible with GEDCOM, but with the addition of a structured extensibility mechanism. It will be named the Extended Legacy Format, or ELF for short. We will also take the opportunity to clarify GEDCOM’s ambiguities while specifying ELF.
With both citation elements and ELF, we are working towards a suite of standards, with each focusing on one self-contained aspect of the subject area. This provides greater modularity, and will allow development of each to progress at its own pace. To date we have issued ten drafts of six different standards in these two areas, with several more drafts under development.
Key to our vision is that the main parts of a component standard must not depend on any specific host language so that new host languages can readily be added; nor can the host language standards know about every component standard they might host, else they would not be properly extensible. Instead there will be a separate lightweight standard or binding explaining how the component standard is used with that host language. We believe this is critical to ensuring maximal interoperability between our standards and third-party standards, whether open or proprietary; it also helps future-proof our work.
Development of a suite of citation elements standards began in 2016 and we released the first public drafts of two standards in June 2017. As a result of the useful feedback we received, we issued updated drafts of both later that year. We currently expect there to be three more citation elements standards: one will define a vocabulary for identifying common citation elements, while the other two will provide bindings for GEDCOM X and ELF. The ELF bindings will be an important test of ELF’s extensibility mechanism, and developing that mechanism was the major focus of our technical work in 2018 and will continue to be in 2019.
When we began work on ELF, it rapidly became clear that, even though ELF was not to depend on citation elements nor vice versa, there were many low-level facilities in common between them. We also felt it necessary to have a clear theoretical foundation to the ELF extensibility mechanism so that extensions can interoperate smoothly. To achieve this we spent the first quarter of 2018 working on three low-level standards defining these low-level facilities and a discovery protocol which will allow applications to obtain information on unfamiliar extensions. These were released as first public draft in March, and at the same time we updated the two current citation elements drafts to build on these facilities.
While this low-level material is not the most exciting, it serves a crucial role in providing a solid underpinning for future work and has allowed us to make more coherent progress on ELF. It is unfortunate that we were unable to engage with members and the public on this important work, and received no feedback on it, either positive or negative, as we rely on feedback from our members to know whether they are content with our direction. However, even with hindsight, we are unsure what we could have done differently to have generated public engagement at this stage.
Since March we have focused our attention ELF. Early on we decided to structure the standard in two parts: one covering the serialisation format, the other on the data model. The new extensibility mechanism works at the serialisation layer and maps tag names to IRIs which define the purpose of the ELF structure. We are calling this mapping a schema. We are also adding a new Unicode escape mechanism, and a way of tagging lines with the language and datatypes of their payloads. Specification of this is well under way, and we believe we have solved the main technical difficulties. We hope to release a first public draft of the ELF serialisation standard in the coming months.
There is much new functionality that could potentially be added to ELF data model, but for version 1.0 of ELF, we have limited our scope to resolving ambiguities in GEDCOM and recognising where current practice is at odds with the GEDCOM standard. This will inevitably disappoint some, but we do not want ELF 1.0 to be too ambitious, lest we never complete it. The most important part of ELF 1.0 is its extensibility mechanism, and when that is delivered it should be easier to issue updates to ELF data model in an interoperable manner. A first draft of the data model standard is largely written, but will need updating to match the terminology used in the serialisation draft.
While drafting the ELF data model standard, it became apparent that GEDCOM’s only complex datatypes were all time-related, and that together they formed a self-contained subsystem. We therefore decided to move these to a separate document which we released as a first public draft at the end of December. The comments we have received so far has been helpful and positive; this feedback will be used to update the draft in due course.
Developing ELF has been slow. To some extent this has been due to unforeseen technical difficulties which have taken time to discuss and resolve, but largely it has been from lack of development resources. We have tried several approaches to development, starting with our call of papers initiative, followed by open discussion on our mailing list, and setting up technical groups made up of members to work towards standards, but the only approach which has produced results is when the Technical Standing Committee (TSC) has done the initial development and released first public drafts for discussion. This means all our standards are currently being written by the two members of the TSC, both of whom are volunteers with full time jobs unrelated to genealogy, which has limited the speed of development.
We are keen to spread this workload to a small team working under the TSC. If you have the expertise and can spare the time to be actively involved in writing detailed technical specifications of the various parts of our ELF and citation element standards, do please contact the TSC to discuss this, as one or two more people contributing high-quality standard text to the project could allow us to progress much faster.
Governance
The board met electronically eight times in 2018, with meetings normally scheduled every six weeks. The minutes of these are publicly available on our website. Board meetings always include a report on how technical work is progressing, and we often discuss our technical objectives and longer-term vision; often there are routine administrative or financial matters to be agreed too. The other item of business that has frequently been on our agenda is reform to our governance structure.
FHISO grew out of the BetterGEDCOM initiative when it was recognised that a more structured environment was needed for technical progress to be made, and after some of the large vendors explained that they would find it difficult, politically, to accept or contribute to third-party standards unless they were developed by an organisation with a formal legal identity of its own. This lead to FHISO being incorporated as a non-profit organisation in Arizona back in 2012, and even though incorporation in Arizona involves significant on-going expense, we believe it remains in FHISO’s best interest.
Incorporation requires us to set out our governing rules in a set of bylaws. With hindsight, our original bylaws were overly complex and better suited to a much larger organisation. As a small team of volunteers, we found it hard to operate within this framework, and often found ourselves diverting effort from technical work to more bureaucratic activities. In an attempt to resolve this we have been steadily reviewing and simplifying our bylaws to make FHISO a more dynamic and responsive organisation, and this process continued in 2018.
Although FHISO is a non-profit organisation in Arizona, it does not currently have tax-exempt status from the US federal government. We believe achieving this would simplify aspects of our financial position, and so we hope to make progress towards a 501(c)(3) application in 2019. The first step of this is to finish updating our bylaws so that they can be reviewed by the IRS as part of the application.
Eight people currently serve on the board of directors and we have all been in office for between four and six years. Four of the eight are currently taking a break from FHISO activity due to changes in their personal circumstances, but so far we have chosen not to seek replacements as we believe eight directors is more than we need. Therefore, the last major reform we are planning to our bylaws is to give us the flexibility to reduce the size of the board. We aim to complete this process by the summer and then run elections for the board of directors.
Some of the current directors have expressed willingness to run for re-election, but we also feel it would be beneficial to have some new people with their own new ideas and insights on the board. We also believe it is important to give you, our members, the opportunity to have your say on how the organisation is run. To thrive, a volunteer-run organisation like FHISO needs a diverse and enthusiastic board who can develop a coherent and viable vision for its future. Do please consider whether you would like to stand for election and be part of shaping our future.
Finances
Summary accounts for 2018 have been approved by the board and published on our website. Accounts for each previous year FHISO has existed can also be found there. The recurring cost of running FHISO currently stand at around $600 per annum, with the majority of this being spent on a virtual mailing address and registered agent in Arizona, as required by state law. We believe there is scope for reducing these costs appreciably, but this will be more easily done after we have completed our 501(c)(3) application. Our next largest expense is the service charge for our bank account which totals $180 per annum.
In some years, though not in 2018, we have paid for professional advice on aspects of our legal status and governance structure. Regrettably, some of this expense could have been avoided had things been done differently in FHISO’s first few years. We anticipate further such expense in 2019 as we progress our 501(c)(3) application. Our current estimate for this expense is $500, though it may overrun if there are complications.
FHISO’s income comes entirely from membership dues and donations. When totalled over the lifetime of the organisation, donations exceed membership dues by a large margin. A large majority of the donations received have been from a small number of our members, often given to cover specific costs. However, relying on donations from a small number of members is ultimately unsustainable, and risks being undemocratic if it results in those members having greater influence over the organisation.
Most of our members joined in 2013 in the months after we first invited people to become members of the organisation. The $20 membership fee has always been described as an annual fee, but we felt it inappropriate to charge membership renewals while FHISO established itself and started technical work in earnest. We believe FHISO is now well established, notwithstanding the further changes we wish to make to our governance structure, and with public drafts of six standard now available, technical work is well under way too. With elections to the board planned for this year, we feel the time has come to start charging membership renewals on an annual basis.
Regrettably, it has taken far longer than any of us anticipated to reach this stage. With hindsight, we have made mistakes along the way and could have done things differently to have reached this point here sooner, and for that we must apologise. While we have been establishing FHISO, the need for organisation which is actively engaged in the production of open standards has grown, and we believe we are the only organisation currently doing this.
We hope you continue to believe in FHISO’s objectives and are willing to support our on-going development of open genealogical standards by renewing your membership.