Hacking systems instead of software

[Brooke Ganz]

Hey, so I know this isn't *exactly* a software-related update, but I did just want to drop a quick line to let people here know about a new project I'm involved in. Last week, I became (I think) the first genealogist ever to try to use a state Freedom of Information law to compel the release of genealogically important records back into the public domain.  I filed a FOIL request back in January with the New York City Municipal Archives, and after first being approved and then denied and then appealing and having my appeal denied, I ended up filing what's called an "Article 78" Petition in court last week in New York.  My attorney thinks we should have some kind of resolution to the case probably by the end of the year.

You can read the background on the FOIL story here:

http://www.avotaynuonline.com/2015/09/public-records-access-one-genealogist-can-make-a-difference/

And here's the new group I just started to help other people learn about FOIL requests for genealogy data:

http://www.reclaimtherecords.org/

I bring this up here, though, to underscore how important it is for there to be new open source ways to publish genealogical data sets. If I win the case, I'm going to suddenly have tens of thousands of never-before-public vital records on my hands.  And while the Internet Archive will be happy to host the microfilm scans, and I'll also be happy to share the films with as many organizations as may want copies (yes, I've already approached FamilySearch), the reality is that this data will most likely get grabbed and posted behind paywalls at one of the for-profit companies, or at least any future transcription of it will end up behind a paywall.  They simply have the skill and size to do it quicker and pay for and manage the servers.  While I guess having it online at a for-profit site is better than not having any access to the data at all, which is the current situation, it still makes me sad that there isn't an easier way to get this kind of genealogy data put up quickly in a free and searchable way as a first resort.

We desperately need more open source tools to (1) upload, manage, and flip through scanned images in discrete sets, (2) transcribe information from the images into spreadsheets/databases, and (3) make those data sets searchable and public -- or else more and more of our heritage is going to end up being the literal property of whomever can martial the resources to cut exclusive deals with state archives and/or stick it all behind their paywall and/or has the technical skill to run the servers.  And right now that's all in the hands of, like, five companies worldwide.

Speaking of which, I highly recommend people take a look at the multitude of comments on this recent blog entry from NARA:

http://narations.blogs.archives.gov/2015/07/31/ancestry-com-partnership-agreement-for-public-comment/

Don't get me wrong, I've been a customer of all the major for-profit sites for years and years, and they're very useful for research.  But as developers, we also really, really need to focus on making open source tools owned by the community, available to everyone, without paywalls -- and right now I'm not seeing much development going on in that regard...

- Brooke Schreier Ganz

[Colin Spencer]

Brooke

Good luck with the case. It is a pity that you have to go to such lengths to get access to the records.

[Tony Proctor]

In the UK, after having a number of polite enquiries being "fobbed off|", I resorted to a FOI request. Unfortunately, the information I was after had not retained, but I wouldn't have known that otherwise.

 

Good luck!

 

    Tony Proctor

--

---
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.

[Robert Hoare]

Brooke,

Well done for chasing it up this far.  Especially given the costs involved. Do you have any idea yet how much the image scanning will cost? Are you considering crowdfunding the costs?

You're going to have more than "tens of thousands" of vital records.  The earlier collection https://familysearch.org/search/collection/2240282 seems to have 120,000 births a year around 1908.  

With the population growth, it'll reach 180,000 births a year by 1929. So 20 years of an average 150,000 births a year is three million births (nearly 9 million named people). Possibly 50-100,000 index page images. Not "big data", but not tiny either.

Internet Archive is one place to archive the whole reels (be sure to add a CC license, to override their default "scholarship and research purposes only" access restriction), but there doesn't seem to be a way to link to specific images (pages).  

Would Flickr (or another image hoster) be more suitable for your point (1), with each reel in an album? They already have many large CC0 image archives (and have an upload API). When the images are individually addressable, and assuming they're in some logical order (date?), a first-pass index could list (for example) the starting date of each page, and link to each hosted image. Basically a table of contents. 

Do you know if the indexes are handwritten, typewritten, or typeset?  Either of the latter two could be OCR'd - not fully accurately, but as a quick way to maybe get an easy 80% full-text index (like a newspaper archive index).

Ben Brumfield's FromThePage looks really nice as a transcription system (crowdsourced), but it doesn't seem ideal for tabular data. Pybossa's free crowdsource-as-a-service platform Crowdcrafting.org might be more suitable, as they already have a crowd to start it off.  These might cover your point (2) (but I agree something simple for tabular data is needed, it's an area I'm looking at as I also need it).

Maybe a commercial company could do transcriptions faster (it shows it the poor results sometimes!), but it's always good to have multiple independent transcriptions (ideally double-keyed, and with an open license).

For (3) - make the data searchable and public - is I think the easiest.  Your transcribed index would be small (a few hundred MB), so assuming the image are hosted elsewhere any small server would do the job, with a simple database browse.  Even just a bunch of downloadable csv's (split by first letter of last name, or by year) would be a start, so that anybody can then make the data available how they like.

Rob

On Sunday, September 6, 2015 at 6:34:02 PM UTC-7, Brooke Ganz wrote:

Hey, so I know this isn't *exactly* a software-related update, but I did just want to drop a quick line to let people here know about a new project I'm involved in. [...]

We desperately need more open source tools to (1) upload, manage, and flip through scanned images in discrete sets, (2) transcribe information from the images into spreadsheets/databases, and (3) make those data sets searchable and public -- or else more and more of our heritage is going to end up being the literal property of whomever can martial the resources to cut exclusive deals with state archives and/or stick it all behind their paywall and/or has the technical skill to run the servers.  And right now that's all in the hands of, like, five companies worldwide.

[...]

[Brooke Ganz]

Well done for chasing it up this far.  Especially given the costs involved.

Filing FOIL requests is free!  Even FOIL appeals!  And all the legal help I received from the NY Committee on Open Government was free too; their staff time is paid for by the NY State Legislature.  The only cost I'm facing so far is the cost of the attorneys, who are frankly giving me an extremely good deal, because they don't think this case will take that many billable hours, and because they seem to genuinely like going up against recalcitrant government agencies in NYC FOIL cases.  (They often have to deal with the NYPD, whom as you might imagine are terrible about records access.)  When I walked into their office and was waiting for my initial meeting, I noticed that their WiFi name was "The Justice League" and I thought yup, I'm in the right place.

And New York State's FOIL has a provision that if a requestor wins his/her case and the government agency really had no good reason to deny his/her request (i.e. it wasn't a good faith disagreement over some legal interpretation of the records access), then the requestor can recoup his/her attorneys' fees!  That's more commonly awarded in other jurisdictions in New York State outside of New York City, but it still might be possible to do if I win. I'm not counting on it, though.

Do you have any idea yet how much the image scanning will cost? Are you considering crowdfunding the costs?

Luckily, image scanning of the 48 microfilm rolls will be free, albeit tedious, thanks to the fact that the Internet Archive has a microfilm scanner that I can use in off-hours, and the lucky fact that I live within twenty minutes of their headquarters.  If I instead go with a professional microfilm scanning company that will automate breaking down the films image-by-image, the cost would probably be about $3000-5000.  Not unthinkable for crowdfunding, but really, I don't want to be touching other people's money at all, if possible.

You're going to have more than "tens of thousands" of vital records.  The earlier collection https://familysearch.org/search/collection/2240282 seems to have 120,000 births a year around 1908.

Um, those are births; the data set I'm going after is an index to marriages.  :-)

Internet Archive is one place to archive the whole reels (be sure to add a CC license, to override their default "scholarship and research purposes only" access restriction), but there doesn't seem to be a way to link to specific images (pages).  

 

Exactly, they are basically just a free server to use, no matter how big the data is (1 TB? more?) and certainly "better than nothing".  But their interface is not great for this purpose.  As I lamented, the genealogy community unfortunately lacks a proper central location for hosting genealogically relevant image sets.

Would Flickr (or another image hoster) be more suitable for your point (1), with each reel in an album? They already have many large CC0 image archives (and have an upload API). When the images are individually addressable, and assuming they're in some logical order (date?), a first-pass index could list (for example) the starting date of each page, and link to each hosted image. Basically a table of contents. 

Huh.  That's an idea!

Do you know if the indexes are handwritten, typewritten, or typeset?  Either of the latter two could be OCR'd - not fully accurately, but as a quick way to maybe get an easy 80% full-text index (like a newspaper archive index).

Handwritten, booooo.  They are supposedly arranged in many books divided by borough (county), then by year, then alphabetically (by first two letters of surname), and then broken down so that grooms are one side of the page and brides on the facing side of the page.

Ben Brumfield's FromThePage looks really nice as a transcription system (crowdsourced), but it doesn't seem ideal for tabular data. Pybossa's free crowdsource-as-a-service platform Crowdcrafting.org might be more suitable, as they already have a crowd to start it off.  These might cover your point (2) (but I agree something simple for tabular data is needed, it's an area I'm looking at as I also need it).

I think I found something that might be perfect, but I still haven't gotten a straight answer as to whether it's open source or not, and whether other genealogy groups can use it.  Check out the State of Washington's wonderful new "SCRIBE" record transcription system:

https://scribe.digitalarchives.wa.gov/

It even has leaderboards!  Well done, Washington!

Maybe a commercial company could do transcriptions faster (it shows it the poor results sometimes!), but it's always good to have multiple independent transcriptions (ideally double-keyed, and with an open license).

Definitely. And my guess is that one of the for-profits will end up grabbing a copy of this data and transcribing it, but then stick the transcription behind a paywall.  *sigh*

(Yes, of course I have been in touch with FamilySearch -- they were the first group I thought of telling about this project!  I would love it if they put this data online for free for everyone. But for various reasons, the specifics of which I don't want to get into here on a public list, I was told that they may not be able to ever post copies of this data online, even though it's public domain data. Let's just say that the NYC Municipal Archives have a really bad reputation of acting like irrational, greedy jerks when it comes to open data.  E-mail me privately if you want details.)

For (3) - make the data searchable and public - is I think the easiest.  Your transcribed index would be small (a few hundred MB), so assuming the image are hosted elsewhere any small server would do the job, with a simple database browse.  Even just a bunch of downloadable csv's (split by first letter of last name, or by year) would be a start, so that anybody can then make the data available how they like.

If only someone made an Apache Solr driven system that can do that (and won second place in the 2012 RootsTech competition for it):

http://www.leafseek.com/

:-)

Seriously, though, the only two sites I know of that use LeafSeek so far -- the Israel Genealogy Research Association's "All Israel Database" and Gesher Galicia's "All Galicia Database" -- are closing in on a combined total of just shy of one million searchable records...and should hit it by end of the year, wooo!

But they're both one-off customized systems.  Until there's a standalone multi-tenant system somewhere, the broader genealogy community still lacks an easy way to spin up new searchable databases for any newly available data sets.  And that kinda stinks.

- Brooke

[Ben Brumfield]

If the scans are uploaded in a manner conformant with their persnickety naming conventions, the images should be available through the image API.  I've used this for about six years in my integrations between IA and FromThePage, and will be happy to help out on that front.

I will be participating in an IIIF hackathon in Philadelphia at the end of this month, and am told that someone from IA will be there as well.  I'll ask them about plans for IIIF support and mention this project.

If any of you are interested in participating, the event page is here: http://www.eventbrite.com/e/iiifification-hackathon-2015-registration-17750618578

(Lots of interesting stuff is happening with IIIF which may concern the genealogy world.)

Ben

[Ben Brumfield]

I've got a bit more time now, so can comment more coherently.

On Monday, September 7, 2015 at 8:26:26 PM UTC-5, Brooke Ganz wrote:

Internet Archive is one place to archive the whole reels (be sure to add a CC license, to override their default "scholarship and research purposes only" access restriction), but there doesn't seem to be a way to link to specific images (pages).  

 

Exactly, they are basically just a free server to use, no matter how big the data is (1 TB? more?) and certainly "better than nothing".  But their interface is not great for this purpose.  As I lamented, the genealogy community unfortunately lacks a proper central location for hosting genealogically relevant image sets.

I wrote above that the Internet Archive actually manages this really well, though the documentation is strangely labeled.
See the Book URL docs: https://openlibrary.org/dev/docs/bookurls

Based on that, if we look at a census reel that's been uploaded and made it through the IA "Derive" task-- https://archive.org/details/populationschedu1370unix -- but we only want the first page of the Pittsylvania County entries, which start halfway through the book, we can manually navigate to the first page using the Bookreader and find it at https://archive.org/stream/populationschedu1370unix#page/n161/mode/1up .  But we don't want to use the Bookreader -- we want to directly link to the URL for the image.  We can do that by converting the URL of the Bookreader, replacing "stream" with "download", replacing "#" with "/", and replacing "/mode/1up" with ".jpg".  That yields https://archive.org/download/populationschedu1370unix/page/n161.jpg and if we visit it, we are redirected to the actual (non-persistent) URL for the image.

There are alternative mechanisms for presenting images from the Internet Archive beyond deep-linking.  When I built the Internet Archive integration for FromThePage, I modified the Bookreader javascript code  (actually the server-side code that sets initializes the javascript) to only display a single page, and turned off the navigation controls.  That let Bookreader handle the URLs for panning and tiling.

Note that you'll need to ingest the IA metadata for the book into your transcription system.  This is non-trivial, as you'll need to read the scandata.xml and possibly the djvu.xml files produced by the derive process.  I've got code that does this at https://github.com/benwbrum/fromthepage/blob/master/app/models/ia_work.rb#L99 -- currently released as AGPL, but I can switch to a different license if you like.

Ben Brumfield's FromThePage looks really nice as a transcription system (crowdsourced), but it doesn't seem ideal for tabular data. Pybossa's free crowdsource-as-a-service platform Crowdcrafting.org might be more suitable, as they already have a crowd to start it off.  These might cover your point (2) (but I agree something simple for tabular data is needed, it's an area I'm looking at as I also need it).

Robert is correct -- I wouldn't use FromThePage for structured tabular data like this.  It looks like I'll be building support for tabular data into the tool in the next few months, but that will still be optimized for incidental, variable-format tables embedded into longer textual material, like specimen tables within naturalist field books or incidental financial accounts within miscellanies (e.g. https://archive.org/stream/Jeremiah_White_Graves_Diary_Volume_2_Book_01/JWGravesVol2Book01#page/n16/mode/1up ).  However, the discovery and delivery system is totally unsuitable for genealogy searches, and the encoding will require much more effort from transcribers than is needed in a fixed-format transcription tool.

I'd look into Crowdcrafting, but also think that a lot more is in the works.  Certainly a general-purpose structured-transcription/search-DB tool is what Ben Laurie and I set out to build a few years ago, but we postponed development on the transcription side in the middle of 2013.  Mainly that was to switch focus to the search database we produced for FreeREG2, but partly that was because the NEH awarded a $300,000 grant to NYPL and the Zooniverse to do exactly that -- enhancing the Scribe codebase that emerged from Old Weather and What's the Score at the Bodleian and was customized by NYPL for Ensemble.  I saw a demo of the tool this May, and understand that while it was to be released eight days ago, it might be a bit behind schedule.  At the same time--and possibly through related work--the Zooniverse folks continued to develop Scribe 2 for the Tate's AnnoTate, released seven days ago, and a similar project at the Folger to be released soon. 

I'll be attempting to touch base with the folks at Zooniverse and NYPL in the next few weeks, as Free UK Genealogy reviews the new systems to decide how to proceed.  Of course, the folks on the other side may be a great deal busier than I am.

Regarding Washington State's Scribe--which is unrelated to the Zooniverse Scribe--I'd also like to know more.  I gather that it was the product of the previous State Archivist, so it may be undergoing transition as personnel and priorities change.  (NB I have no insight into state politics here in Texas, much less in Washington!)  If you do find out anything about the status of the software, I'd love to know more. 

Also, although I've said as much in another forum, good on you, Brooke! You're fighting the good fight, and we all stand to benefit.

Ben

[Robert Hoare]

On Monday, September 7, 2015 at 6:26:26 PM UTC-7, Brooke Ganz wrote:

Filing FOIL requests is free!  Even FOIL appeals!  And all the legal help I received from the NY Committee on Open Government was free too; 

Do you still have to pay for the actual copying (physical reels)?  Or does the legal process get them to waive that?

 

You're going to have more than "tens of thousands" of vital records.  The earlier collection https://familysearch.org/search/collection/2240282 seems to have 120,000 births a year around 1908.

Um, those are births; the data set I'm going after is an index to marriages.  :-)

Good point. :-)  I went and read the Avotaynu article again, indeed there are plenty of mentions of marriages, not sure how I thought they were births ...  The data's a lot more complex than simple births.  So there's still a lot of data, say only 50,000 marriages per year (a million total?), but lots more people and places mentioned.

  

Check out the State of Washington's wonderful new "SCRIBE" record transcription system:

https://scribe.digitalarchives.wa.gov/

Does look pretty, although I didn't spot anything about it being open source.

 

If only someone made an Apache Solr driven system that can do that (and won second place in the 2012 RootsTech competition for it):

http://www.leafseek.com/

:-)

I was thinking the NY data was simple birth records (just a handful of columns) so didn't think of full-text search engines, more just super spreadsheets.  These multi-page marriage documents are much more suited to your Leafseek.  (I did take a look at Leafseek a couple of years ago, but I didn't have any data that's suitable for it - and I find Java a pain to manage).

But they're both one-off customized systems.  Until there's a standalone multi-tenant system somewhere, the broader genealogy community still lacks an easy way to spin up new searchable databases for any newly available data sets.  And that kinda stinks.

Could Leafseek do that role, with a bit of splitting things off into parameters for each tenant? 

A lot of other data are simple lists, so a full text search might be overkill for that (and not precise enough, if it mixes up names and places).  So a read-only online spreadsheet with filtering and searching might be enough?

Hosting data like this is the sort of thing genealogy societies should have been taking the lead on, but most of them (especially in the UK) still sell their data in book or even microfiche form and charge for data access.

Maybe what is needed is a global volunteer "Genealogy Records Society" dedicated to the obtaining, preservation and (legal) free distribution of freely reusable genealogy records and data  (without any commercial or religious motivation) ...  does anything close to that exist?  

Rob

[Justin York]

You rock.

[Brooke Ganz]

I wrote above that the Internet Archive actually manages this really well, though the documentation is strangely labeled.
See the Book URL docs: https://openlibrary.org/dev/docs/bookurls

Based on that, if we look at a census reel that's been uploaded and made it through the IA "Derive" task-- https://archive.org/details/populationschedu1370unix -- but we only want the first page of the Pittsylvania County entries, which start halfway through the book, we can manually navigate to the first page using the Bookreader and find it at https://archive.org/stream/populationschedu1370unix#page/n161/mode/1up .  But we don't want to use the Bookreader -- we want to directly link to the URL for the image.  We can do that by converting the URL of the Bookreader, replacing "stream" with "download", replacing "#" with "/", and replacing "/mode/1up" with ".jpg".  That yields https://archive.org/download/populationschedu1370unix/page/n161.jpg and if we visit it, we are redirected to the actual (non-persistent) URL for the image.

Thanks, this info is really, really helpful!  I imagine what I will be doing is scanning and uploading one microfilm roll to the IA system as a test and then asking for advice and comments about the formatting, before going ahead with that format for the subsequent rolls.  Um, *if* I win the files, I mean.  Gotta keep that qualifier in there for now.

Note that you'll need to ingest the IA metadata for the book into your transcription system.

Yes, but we don't have a transcription system yet, so this is still hypothetical. :-)

 

I'll be attempting to touch base with the folks at Zooniverse and NYPL in the next few weeks, as Free UK Genealogy reviews the new systems to decide how to proceed.  Of course, the folks on the other side may be a great deal busier than I am.

Please let us know how it goes!

 

Regarding Washington State's Scribe--which is unrelated to the Zooniverse Scribe--I'd also like to know more.  I gather that it was the product of the previous State Archivist, so it may be undergoing transition as personnel and priorities change.  (NB I have no insight into state politics here in Texas, much less in Washington!)  If you do find out anything about the status of the software, I'd love to know more. 

I'll let you know if they ever write back to me!  I managed to get on their internal e-mail mailing list about it but no one has answered my questions yet...

 

Also, although I've said as much in another forum, good on you, Brooke! You're fighting the good fight, and we all stand to benefit.

Aw, shucks. Thanks!

- Brooke

[Brooke Ganz]

Do you still have to pay for the actual copying (physical reels)?  Or does the legal process get them to waive that?

Nope, I still have to pay for any copies made.  I looked around online and it seems like their offer of $35 per microfilm copy was very reasonable; NARA charges far more than that for copies of their films, although NARA includes shipping in their cost.

(I did take a look at Leafseek a couple of years ago, but I didn't have any data that's suitable for it - and I find Java a pain to manage).

Oh, it is.  But once it's finally up and running and configured kinda-sorta-okay then it's pretty stable.

 

But they're both one-off customized systems.  Until there's a standalone multi-tenant system somewhere, the broader genealogy community still lacks an easy way to spin up new searchable databases for any newly available data sets.  And that kinda stinks.

Could Leafseek do that role, with a bit of splitting things off into parameters for each tenant? 

Someday, I hope.  I've been thinking about it a lot.  In the past two years, Solr has gotten better at accommodating multiple cores and multiple collections within the cores.  But it would still be a lot of dev time to make it happen in a nice automated way, invisible to an end-user who just wants to make an account and upload data sets and have the system spit out pretty search engines of their data.

 

Hosting data like this is the sort of thing genealogy societies should have been taking the lead on, but most of them (especially in the UK) still sell their data in book or even microfiche form and charge for data access.

I know, and I don't like it.  Personally, these days I only ever do genealogy projects (whether coding or transcribing or obtaining records, or whatever) where the end result will be free to everyone.

 

Maybe what is needed is a global volunteer "Genealogy Records Society" dedicated to the obtaining, preservation and (legal) free distribution of freely reusable genealogy records and data  (without any commercial or religious motivation) ...  does anything close to that exist?  

I don't think so.  FamilySearch is probably the closest model in real life, and does the best job of this.  They even tend to look the other way if people scan their films, as long as nobody's profiting from the scans, which is awesome.  And this is all great when they're taking the lead in preserving all kinds of records worldwide, which I love about them, but not so great when they also make trees and data structures that exclude theologically-unpalatable people and relationships, which drives me nuts.

...yeah, an open genealogy records society would be awesome.

- Brooke

[Brooke Ganz]

You rock.

Thanks, and so do you!  :-)

*hi-fives all around*

- Brooke

 

[Brooke Ganz]

 🎉 I WON MY FOIL CASE!  🎉

Read the full details in the inaugural issue of the Reclaim The Records newsletter! http://eepurl.com/bApxuT

- Brooke

[Rob Hoare]

Congratulations, Brooke!  That's a great precedent (even if not a legal precedent) that should help a lot with your next FOI applications.  

 

Hopefully (once scanned - which I realise is a lot of work yet!) it'll be the first of many fully open genealogy image collections - which should in turn drive the development of more ocr, transcription and indexing software. Really good start!

 

Rob

--

 

---

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/rr96Jc7Bgi4/unsubscribe.

To unsubscribe from this group and all its topics, send an email to rootsdev+u...@googlegroups.com.

[Colin Spencer]

Well Done Brooke. Hopefully that will have far reaching repercussions.

[Justin York]

I can't stop thinking about how awesome this is. I'm anxiously waiting for an opportunity to contribute money or time.

--

---

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.

[Brooke Ganz]

On Monday, October 5, 2015 at 6:58:26 AM UTC-7, Justin York wrote:

I can't stop thinking about how awesome this is. I'm anxiously waiting for an opportunity to contribute money or time.

D'awwww, thanks!  No need for monetary donations, but if you ever run into any record sets that are being wrongly withheld from the public, please take our Records Survey and let me know, so I can add it to our ever-growing "to-do" list?

Link: http://www.reclaimtherecords.org/records-survey/

I had no idea how widespread this problem was until people started flooding me with examples of records that they are not being allowed to see.  Sometimes it seems like the local archivists are just ignorant of the law, or perhaps just being cranky, but in some cases it seems like a straight-up revenue grab by certain places, like when you can easily buy a certificate copy but cannot merely see it first (i.e. Cook county, Illinois).

Also, you can file your own FOIL requests too!  I want to make a "how-to" guide for genealogists to start using this method more regularly.  Genealogists actually do use FOIA-with-an-A (the more famous Federal law) all the time, although we don't always recognize that's what we're doing when we get immigration/naturalization records through USCIS or order SS-5's from the Social Security Administration.  But the state-level FOI laws were just not being used, for some unknown reason.  Taking advantage of those laws will open up a whole new bag of tricks for us.

Right now, my focus is will be on fully finishing up this particular "pilot project" FOIL request and getting the images placed online by end of the year, if not sooner. And then I'll do write-ups of everything I've learned and all the mistakes I made along the way.

And then I'm starting 2016 with A LOT of new requests planned.  I am so psyched.  :-)

- Brooke