I have been researching my family for about 15 years.

  • I have used Microsoft Access for about 10 years to record my results.
  • Microsoft no longer supports the version of Access I have.
  • I have no need for Access in my other work, so do not wish to upgrade.

    What things should I look into for a Genealogy/Family Tree program, to ensure my relatives/descendants will be able to use my research in the future?

  • Welcome to Genealogy.SE, Edward. I added some detail to your question title; please confirm that it still conveys your meaning.
    – GeneJ
    Commented Oct 28, 2012 at 7:24
  • 1
    When you say "have used Access" do you mean just the tables, or have you created Forms etc?
    – Andrew
    Commented Oct 28, 2012 at 8:19
  • 1
    @Andrew - I have Forms, reports etc.
    – Edward
    Commented Oct 28, 2012 at 20:04
  • 2
    Can you export as XML or similar format?
    – Those Legs
    Commented Oct 29, 2012 at 11:30
  • 2
    @Andrew - I am also concerned that as I need to upgrade windows, I will reach a stage where my Access will no longer run.
    – Edward
    Commented Nov 3, 2012 at 2:11

10 Answers 10


Perhaps the single most important feature in selecting any software is your ability to extract data from it in a non-proprietary format. Whatever tools you use should have an "export" function that enables you to switch to something else at any time. That means that the file you generate must be able to be read by another (and preferably more than one) program.

There is currently wide-ranging discussion around whether the (one-time standard) GEDCOM is still the most appropriate means to interchange genealogical data. One thing that you might investigate is whether a content-neutral format (such as XML) may be a relatively future-proof option.

  • 5
    Export capability needs to be your primary criterion, and if you plan to migrate in the the foreseeable future, that means GEDCOM. (The sad history of trying to get beyond GEDCOM is outlined here.) Tamura Jones has an overview of GEDCOM which may be helpful if you are not too familiar with it.
    – RobertShaw
    Commented Oct 29, 2012 at 1:19

This is something we should all be worried about. I was so concerned with relying on proprietary software, databases, and data formats that I devised my own replacements a couple of years ago.

I'm not recommending this approach for everyone but we all need a way to represent our data that will be long-lived (allowing it to be passed on), non-proprietary, hardware-neutral, and ideally locale-neutral.

This is why I became an organising member of FHISO. I apologise for this sounding like a simple advertisement for them but that is precisely how they came about. While it doesn't solve your question today, there are some people working very hard behind the scenes to solve it "tomorrow".

Worth checking them out!

  • Worth checking them out "tomorrow", since a new standard will likely be two years or more in the future. Standards almost always take quite a bit of time.
    – RobertShaw
    Commented Oct 29, 2012 at 0:48
  • 6
    Or is Tony suggesting - check them out to see if you can assist? Commented Oct 29, 2012 at 5:49
  • 3
    Support from everyone is essential - assistance would be extremely welcome too though. Coming up with a good design is only part of the problem, and there are already some good designs out there. Unlike previous attempts at a standard (e.g. GenTech & OpenGen), FHISO are not beginning with a standard and hoping everyone will run with it. They're trying to create a representative body for all of genealogy, worldwide - one that will represent our common interests for years to come.... (cont)
    – ACProctor
    Commented Oct 29, 2012 at 14:26
  • 1
    It is therefore beginning by getting all the major stakeholders (content providers, software vendors, societies, & passionate individuals) around the same table. It will be those members that will then do the standards work - together. The existing volunteers will transition to a self-governing organisation very soon.
    – ACProctor
    Commented Oct 29, 2012 at 14:27
  • +1 particularly for letting me know about FHISO.
    – Lisa
    Commented Jul 1, 2013 at 5:31

The key goals to help future proof any research would be:

  • to use software and operating system that is supported and updated
  • to will your computer, software and data to someone interested in maintaining
  • to give hard copies to friends and relatives

The particular software is not as important as it being able to be used in the future.


To me, "future proofing" your genealogical research so that others can use it in the future means just one thing:

1. Document Your Sources

That is it. If you document exactly where you got everything, then other's will be able to use your work. If not, it becomes hearsay and cannot be verified or trusted for future work by others.

So, in choosing a genealogy program, choose one that best helps you document all the source details that become the evidence that lead you to all your conclusions.

Note: GEDCOM is not good yet at transferring source information between programs, and that's something that future standards will hopefully fix. But that should not stop you from recording this information. Even if, your relatives/descendants one day may have to manually reenter your sources, at least they will have them so that they can reenter them.

  • 2
    Right. Source, backup and preferably also print out. Commented Oct 29, 2012 at 12:06
  • 2
    I hope everybody keeps their source material organized in hard copy, or digitized somewhere on their computer outside of their genealogy software. The software is basically to help pull everything together and organize conclusions. But effectively, if the software and its data get destroyed, then all the conclusions should be able to be reconstructed from the source material.
    – lkessler
    Commented Oct 29, 2012 at 13:11

[EDITED to reflect comments]

One way to improve the longevity of your data is to use open-source tools to make sure your data formats are well-documented and that gratuitous changes by software vendors not orphan your data. This is particularly problematic for Software-As-A-Service (SAAS) offerings (e.g., the Ancestry.com site), for software in which at least some important aspect is delivered online, for software distributed through controlled means such as App Stores, and for software that depends on versions of operating systems that are no longer supported.

While open-source software can and does evolve, it is always possible to download or build earlier versions of the system or to modify the source as required. Also, while the majority of users may not have the skills or desire to modify software, there will be others who can and will make the changes or develop the tools. After all, that's often how the open-source software got created in the first place.

With respect to genealogy, on Open Source tool for managing your research is called gramps, and you can download it from http://gramps-project.org/

  • Gratuitous changes may very well orphan your data with regards to version upgrades anyway, although it's less likely. But you are right with regards to the openness of the data and in such cases it will be possible to get help to export the data, at the very least. Commented Oct 28, 2012 at 17:53
  • The nice thing about open-source sw is that you can always go back to an older version, and it is easier (compared to commercial sw) to write tools to transform data in sensible ways. Commented Oct 28, 2012 at 18:00
  • 1
    This is difficult for less technical people. And older versions often have bugs, less features, and may not even run on newer operating systems.
    – lkessler
    Commented Oct 28, 2012 at 20:23
  • Well, you can always go back to an older version with commercial software too. But otherwise I agree. Commented Oct 28, 2012 at 21:23
  • 1
    @LennartRegebro the challenge is that if a proprietary piece of software does not implement a particular piece of functionality (e.g., export to SuperDuperGenealogyInterchangeFormat), you're stuck. With Open-source software, there is a good chance that someone could write that tool. Commented Oct 28, 2012 at 21:26

Another thing to bear in mind is not only the data format, but the medium upon which it's stored. Electronic media such as data CDs or DVDs have unknown lifespans - data CDs are falling out of use now after about 25 years (although they'll clearly be supported by hardware for some time to come), and the same will likely happen to DVDs as Blu-Ray becomes cheaper and more common. Devices like USB flash drives will survive only as long as USB remains in common use (probably awhile yet). As with photographs, all the aforementioned need to be protected from excess heat and humidity. The somewhat ironic reality is that the medium with the longest likely life expectancy is still paper.

The long-term survival of the data is also somewhat proportional to the number of copies that exist, regardless of the medium. So best bet is use all of the above, and make multiple copies in each new medium that becomes available.


Like any data that you want to "future-proof", one good guideline is that it should be human-readable. One key component of this is that it should be in text rather than binary. XML and GEDCOM are both examples of human-readable formats, although XML is much more readable.

  • I think that's the first time I've ever seen XML referred to as "readable", much less "more readable". (Though I'm not disagreeing)
    – Crwth
    Commented Oct 29, 2012 at 12:50
  • Have you tried to read a GEDCOM file? :)
    – Jeremy
    Commented Oct 29, 2012 at 14:38
  • 1
    Like a gripping novel... but it sure makes for writing an easy parser!
    – Crwth
    Commented Oct 29, 2012 at 20:05
  • Good point. And if you you find the readability of XML flavours to be wanting you can easily (if you do the odd programming) write a parser whereas any binary format will require special skills and far more guesswork to write a parser 100 years hence.
    – Lisa
    Commented Jul 1, 2013 at 5:23

The answer probably depends on what you currently have in access. If it's just persons and families, it is not that difficult to find a way to export those to a program like Gramps, which has a nice CSV import, and use GEDCOM if you later want to change to another program. The Gramps CSV format is well defined and can also be expanded if you wish.

If you have tables that contain detailed information from sources that are more difficult to convert, I suggest that you first export your data to a format that is human readable, like CSV, or portable like SQLite, so that you can explore what you have with modern tools. I recommend SQLite because it is supported on platforms like Windows, Linux, even Android, and it gives you a chance to save your own design, and wait for standards to evolve in a direction that supports the way you like to work.

For what is worth, SQLite is also used in RootsMagic, so if you know what to do, you can probably hack that too.


Another good method of future-proofing is to post your familytree information (about non-living people) online in a variety of places, for example: WeRelate.org and WikiTree. I also post much of my genealogy research on my blog and then regularly save those posts at the Internet Archive (which is like a free online archives for web pages).


There have been some great responses on this already but I have not seen a couple key points mentions so I will put out what my plans are that I have specifically outlined.

  1. I make lots of backups of my data in multiple formats, both in the program I use formats as well as export trees to text and word formats. You also need to back it up to multiple accessible means. This still does not get past the data format issue though related to both inconsistently supported Open Source software or Proprietary ones.
  2. I strongly agree XML is great and someone will always be able to parse it, but under what XML schema is really becomes the question. CSV is at least halfway readable by a human, but the classic print format is definitively readable. The LDS church is a primary subject matter expert and probably the largest repository on the topic at least in the US and so to some degree everyone else is at the mercy of what they are willing to do. That all said, if there were an ISO (International Standard Organization) format is perhaps adopted for at least key data point formats would at least be 'generally' compatible and not as proprietary as someone will always be making their own.
  3. At some point in my life I plan on self publishing a hard bound book with my family history in it to ensure its availability. I then plan to distribute this book to key areas of the country's historical societies where my family originated as well as to major genealogical organizations and libraries. It will not only include my tree, but photos and other clippings as people want to know their ancestry, but they also want to hear the stories. I will also likely publish it in multiple e-book formats.
  4. The hardest thing for me will be to be happy enough with my work and its completeness that I plan to upload it in its entirety to 'the cloud'; both Ancestry.com and FamilySearch or another. It is actually outlined that if I pass before I do this, that it be done after I pass. I have not done this now I have not done so as feel the research is 'mine' to a degree and I am frustrated enough with the quality of data and lack of source or proof and 'guesses' out there that I do not want to contribute to it until my data is clean and I make some key connections, or I know I can upload it and there be a way where I know that my comments and supporting information will be retained.

A thing to note about my tree is I don't just research my direct line, but my tree goes forward back and back forward and I add to each branch until it ends or I hit a stopping point so the tree is quite large which creates other issues that I will have to account for in at least item #3.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.