Family History Developer Conference Salt Lake City Aug. 21st and 22nd

[Gordon Clarke mz]

Fellow Programmers,

I need help planning and preparing for this conference.  I've got the LDS Business College booked for 3 days if we need it.  I'm looking to send out "Call for Papers" in a week or two.  This is a community effort by programmers for programmers.  FamilySearch will be a contributor but this is not a FamilySearch nor RootsTech event.   I would be happy with 150+ attendees and 24+ meaningful session with very little overlap so people are not stuck with always having to choose one class or track over the other.  A smaller conference makes it easier to network and collaborate with peers.   

Please send in:

COMMENTS

ACTIVITY IDEAS

WILLINGNESS TO ATTEND

WILLINGNESS TO HELP MAKE THIS SUCCESSFUL

Should we call the event:

Family History Developers Conference?

RootsDevCon?

Programmers4Families?

???

Best regards

Gordon Clarke

Family History Software Advocate

and

Familysearch Business Development Manager

801-240-0770 office

801-253-6365

[Gordon Clarke mz]

test response

[todd.d....@gmail.com]

I'm very excited for this! It help that it's going to be held in the adjacent building to my work. I'd be interested in presenting a couple of lightning talks. Are we open to that kind of presentation format?

I like RootsDevCon. I'd be willing to design promotional materials/logos/etc.

Cheers!

test response

--

---
You received this message because you are subscribed to the Google Groups "rootsdev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to rootsdev+u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

Tod Robbins

Digital Asset Manager, MLIS

todrobbins.com | @todrobbins

[Dallan Quass]

Me too! I'm moving to Utah this summer so I'm going to be spread thin, but I'm planning to attend and would love to help you make it successful.

[Laura Cutler]

This sounds great! I'm excited to attend.

In addition to some sessions, it might be fun to code something together. If someone has a project they want ideas for or help with, maybe we could divide into small teams and put together different pieces of a solution. That could be good for both networking and idea sharing, and hopefully the sponsor would get some useful ideas out of it.

[Justin York]

I plan on being there.

Are the presentation expected to be 45 - 60 minutes? Will presentations be required to relate directly to genealogy? In other words, will generic "intro to language/technology X" presentations be allowed?

I'm willing to help where I can (not sure what kind of help you need). I would also be willing to present about the architecture and design decisions behind Find-A-Record.

Laura's right; we really need some sort of hackathon. I like the idea of a sponsoring organization, especially if they have an API. Some examples:

* MyHeritage sponsors the contest with the goal to build the most unique/useful/difficult app built on top of their API

* FamilySearch gives each group temporary access to the production API (or loads tons of real data into sandbox) and sponsors it with their API (especially if the record matching API is released before then, or if we can get early access to it)

* Setup a small framework for traversing the tree following the visitor pattern and give us a few hours to come up with the best implementation for the visitor

Or the hackathon could be less specific. I've never participated in a hackathon, so I don't really know what works and what doesn't.

[Brooke Ganz]

This conference sounds awesome, and I'd really, really like to be there -- but I won't know for sure if I can attend until maybe the week before it starts, so I guess count me as a "maybe".

I would like to echo the call for us all to team up to work on coding projects, whether it's a formal hackathon or just "here is a list of things that I wish existed in the genealogy world, but they don't exist yet, so let's build them".  And there are so very many things that we could all add to that list…!

For starters, we're talked for something like two years now about setting up Dallan's Place Names server as a standalone public REST service.  Maybe we could finally make that happen?

Or what about working with various open genealogical datasets to make them available as searchable sites to the wider public, without the middleware of a for-profit company selling access?  I've been working on setting up a project to do this with the New York State 1957-1964 death records, but frankly it's sat half-finished on my server for months now.  There are many similar free and open vital records sets (usually in CSV or Excel format) available from many of the states, counties, and cities, so we can each pick whatever works with our own family history or research interests.

Or why not build tools to analyze, rather than search, through community record sets?  Given a CSV file of thirty years of births for Anytown, can we design some data analysis tools with a simple web interface to display the data in prettier form, rather than record-by-record?  Can we use d3.js to graph births by time, or separate our by gender, or show interconnections between surnames, or clusters of in-migration or out-migration?  Can we build things that will be able to put family history in context with local history and local demography?

Or how about building an API for the Social Security Death Index (SSDI)?  I have five copies of the Death Master File (DMF) from different dates over the past seven years or so, one purchased (legally!) for about $37 through a data broker and the other four copies downloaded from people who had bought the file and then made it freely available online out of their own generosity.  None of these is the latest, most up-to-date version of the DMF, and of course we all know that access to the latest version is going to be restricted starting this year (boo, hiss).  But the old data is not being removed or censored, and aside from the occasional zombie or mis-keyed record, none of the people listed in the old files is going to un-die anytime soon, so why not build an API out of the data so we can all query it as much as we like?

And that's just the tip of the iceberg…there are a LOT of things that we all could be working on!

- Brooke

[Brooke Ganz]

(My previous reply didn't go through for some reason, so trying to cut and paste it again here…)

[Colin Spencer]

Brooke,
 It was marked as Spam by Google. I have just been notified of it in the queue but as you have reposted I have deleted it.

Colin

[Ben Brumfield]

I actually cleared it for posting after getting confused about whether the original had gone through.

Ben

Sent from my iPhone

[Robert Hoare]

On Sunday, March 23, 2014 7:10:31 PM UTC-7, Brooke Ganz wrote:

I would like to echo the call for us all to team up to work on coding projects ... how about building an API for the Social Security Death Index (SSDI)?  

That's a great idea Brooke.  It's so much easier to work on projects when there's some feedback (and some pressure to get it done!).  Like you, I have some partly-finished projects in this area, that I've never found an incentive to finish (too much real work gets in the way).

I'll try to get my copy of the SSDI online with a simple API asap (I have bits of it already done) to see if there's any interest.  I only have one from after the "last place alive" was removed, but if that works OK we could find a way to use one of your more complete earlier copies. 

For the US, the SSDI (even though very incomplete) is a key file (and is also one of the largest).  An interesting project would be to cross-link it with the several different military burial record sets (which are also public data).  And even that New York state death index.

Rob

[Brooke Ganz]

I'll try to get my copy of the SSDI online with a simple API asap (I have bits of it already done) to see if there's any interest.  I only have one from after the "last place alive" was removed, but if that works OK we could find a way to use one of your more complete earlier copies. 

Turns out that I have four copies of the DMF, not five.  They are:

- 2010-03-09

- 2010-11-17

- 2011-11-30

- 2013-05-31

Let me know if you, or anyone on list, wants access to them, and I'll give you links to my DropBox folders.  And I'm always happy to accept trades of other versions, too.  Note that they are really, really big files; you can't just open them in Excel, you'll need to ingest them into MySQL or ElasticSearch or something like that.

The fifth DMF file I originally mentioned is one that I know about, but have been unable to decode so far.  In or around 1997, a copy of the DMF (presumably of 1996 or 1997 vintage) was included on a semi-obscure piece of Windows family tree software called Generations Deluxe version 4.2.  Someone (not me) ripped the CD's and put them on various Torrent sites with, apparently, the DMF included too.  But I have so far been able to unwind the DMF data from the program data.  You should be able to find the file on PirateBay, if you're interested.

Older versions of the DMF are valuable because some data has been stripped from the more recent releases.  For example, older versions contain the ZIP code where the final Social Security payment was sent (if there was one), which was very often the deceased person's last residence.  If you cross reference that data with a public ZIP code database, you can select all the John Smith's from Some Town, Some County, and/or Some State, as opposed to having to sort through a zillion of them.  Obviously, this is useful when you're trying to track down a relative or their family.  For someone like me, doing primarily Jewish genealogy, being able to sort out common names by who did or did not move to South Florida in their old age is often an excellent proxy for ethnicity.  :-)  I would imagine being able to quickly figure out the Utah/Idaho decedents would be similarly useful for other people.

Older versions of the DMF also contain more records in total than even the most up-to-date version does, because on November 1, 2011, over four million records were stripped from the file because they had been sourced from various states who were not supposed to be confirming death data with the Federal government (?!), or some weird rule like that.  Details here: http://blog.familytreemagazine.com/insider/2011/11/02/SSAToRemoveProtectedDeathRecordsFromDeathMasterFile.aspx   So having record sets that predate 2011-11-01 are pretty nice, and would be great supplementary info to any new API that we might build.

 

For the US, the SSDI (even though very incomplete) is a key file (and is also one of the largest).  An interesting project would be to cross-link it with the several different military burial record sets (which are also public data).  And even that New York state death index.

I also have an idea where we just build "API for the Dead Guy", which could be SSDI + all the state/county/city public death indices we can find +  military data + any free cemetery or tombstone data (although Find-A-Grave and BillionGraves are not open data, alas).  Make the API free for developers and for low level usage (<100 requests/day?), but charge in tiers for bulk requests, with the profits (if any) first directed to covering the server and bandwidth costs, and then anything left over donated to non-profit genealogical and historical societies that rescue records and/or put them online.  It's DGaaS -- Dead-Guy-as-a-Service!

- Brooke

[todd.d....@gmail.com]

Alright, here is my proposal for the conference logo with an updated logo design for RootsDev. One PDF is with one accent color and the other is with two accent colors. I'm still playing around with the color palette but I like where it's headed. Let me know what you think.

Cheers!

–Tod

On Thu, Apr 3, 2014 at 3:19 PM, Bob Coret <bob....@gmail.com> wrote:

Great initiative!

I'm tempted to attend, but have to think about the 4978 miles in between The Hague and Salt Lake City....

Idea: let's have the next conference in Europe :-)

Bob Coret

--

---
You received this message because you are subscribed to the Google Groups "rootsdev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to rootsdev+u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Asparagirl]

I like it.  The red/green version is my fave, as long as it doesn't annoy any colorblind members.

- Brooke

Sent from my iPhone

You received this message because you are subscribed to a topic in the Google Groups "rootsdev" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/rootsdev/DLcGtJqaEmE/unsubscribe.
To unsubscribe from this group and all its topics, send an email to rootsdev+u...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

<rootsdev-conference-two-color.pdf>

<rootsdev-conference.pdf>

[Dallan Quass]

I like it as well, though the black-green on is my favorite :-) I'd be happy with either one though.

Thank-you Tod!

[Justin York]

I agree with Dallan.

[Daniel Zappala]

Black and green for me too.

[Luther Tychonievich]

It took me 3 weeks, but I managed to clear out my schedule. Count me in!

[Andrew Hatchett]

Luther, Glad you are going to make it!

Andrew Hatchett

gplus.to/agh3rd

[Luther Tychonievich]

On Thu, Mar 20, 2014 at 1:21 PM, Gordon Clarke mz <myz...@gmail.com> wrote:

Please send in:

ACTIVITY IDEAS

In addition to the hackathon and related ideas discussed already, here are a few other ideas I've been thinking about:

1. Presentations:

I feel like a lot of us are re-designing the wheel over an over again and/or walking into the same dead end that everyone else went down. Thus I propose a set of "community wisdom" type presentations:

1a. The Genealogical Proof Standard and other "good research" practices, and how they can be used to predict limitations of software, inform UI design, etc.

1b. A survey of what's been done (for good or ill): GEDCOM 5.5, Lifelines, the GenTech data model, the new.familysearch collaboration model, …

1c. Intellectual property and "public" records

1d. A survey of visualization techniques with pros, cons, and either algorithm outlines or places we can get the code

… plus, of course, whatever comes out of the call for papers.

2. Panel discussions on

2a. Pros and cons of rolling-your-own, using a public-but-single-party specification, and formalized standards (or: "Do we want FHISO?")

2b. What is/could be/should be the role of machine learning in family history?

2c. How can we serve the professional, the hobbiest, and the prove-I'm-from-royalty dreamers?

3. Working-group type meetings on

3a. Designing (and creating?) a "so you want to be a rootdever" introductory video series/mook/tutorials/wiki/etc.

3b. Creating an "official" rootdev glossary (see http://fhiso.org/files/cfp/cfps74.pdf for my initial effort at such a lexicon, but it just has my opinions in it right now…)

I'm not fully sold on any of these, and some might be better for RootsTech than RootsDevCon. Thoughts?

[Justin York]

+1

--

[Jay Askren]

Has there been any progress on this?  Is there a website with more details yet?

Jay

[Luther Tychonievich]

Related to Jay's comment: Gordon, much thanks for coming up with this idea, booking the venue, and starting us thinking about a conference; do you have the remaining organization in hand or are there tasks some of us might be able to assist with?

--

[rockwel...@gmail.com]

Hi Brooke, 

I was really excited to read your post! I have also been gearing up to build tools to interact and analyze the DMF. I have one of the newer copies which is limited as described and have been searching everywhere for older copies with the broader information set. My intent was to keep the ancillary data intact for records that were not changed or removed over time and for new records that are added after Nov 2011 I would just have the more limited data set. I have been looking into particular state record sets that could be added as a supplement post 2011 and there are a few where that would be possible. My goal is to develop an API once the data set is as clean as I can make it and proper indexed. Is there still a chance you are still willing to provide dropbox access to the old files?

Regards,

Rocky

(rockroland on Ancestry.com)

[rockwel...@gmail.com]

Hi Brooke,

Any chance you were/are able to share the old SSDI data?

Best,

Rocky

On Tuesday, March 25, 2014 at 3:39:27 PM UTC-4, Brooke Ganz wrote:

[Asparagirl]

Hi Rocky,

I'm glad you liked the post. :-) I would be happy to provide access
to the SSDM files I have to you and anyone else who wants them.
They're in my Dropbox right now but I might also upload them to the
Internet Archive (archive.org) at some point, if I can find the time.
But here are the Dropbox links for now:

2010-03-09:
https://www.dropbox.com/sh/urxs2ifssb9oq78/AACSHOilKwsV8xwGVVpX1-nEa?dl=0

2010-11-17:
https://www.dropbox.com/sh/pneyuzakntq8fxa/AABqJCKJ6N-qDo9X4AFDcQxda?dl=0

2011-11-13:
https://www.dropbox.com/sh/hb95kjo3qlnn682/AAAS9UT1ckKukLkIbXI2CcNla?dl=0

2013-05-31:
https://www.dropbox.com/sh/naiq7dqgha8svn0/AACH2RFiu4ZY6oA884NiErnZa?dl=0

I don't have the 1997/1998 one I mentioned in my post, since I was not
able to disentangle the SSDI data from the program data, but the
torrent is probably still up in the usual places, if you want to try
your hand at it.

If you have a copy of the SSDI/SSDMF/whatever and it's not one of the
ones that I have, would you be willing to share it too?

Good luck with your project; I'm happy to see the data get more
widespread use! Are you developing a for-profit product or a free
product or...?


- Brooke Schreier Ganz

> --
>
> ---
> You received this message because you are subscribed to a topic in the
> Google Groups "rootsdev" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/rootsdev/DLcGtJqaEmE/unsubscribe.
> To unsubscribe from this group and all its topics, send an email to

[Rocky Roland]

Thank you so much Brooke! This will really help a lot. I have a late 2013 copy that I will pull together and post a link to. My goal was to make it a free product.. I'll keep you posted.. Thanks again!

Best.

Rocky