JFK Timeline Project

First topic message reminder :

New JFK Timeline Project has been released. Deserves special mention here (apart from how useful it's going to be) because ROKC's very own Steven Duffy had a big part in making it. He gave years of his hard work from his index to the Project. Bookmark this puppy and use it lots!

http://www.jfk-timeline.org/index.php

Paul McGurkenfarklein wrote:Thanks Brian and congratulations!. I'll be sure to go and have a play with it this week.

Paul,

if you could report back with your thoughts and any tips, that'd be great.

greg parker wrote:

Paul McGurkenfarklein wrote:Thanks Brian and congratulations!. I'll be sure to go and have a play with it this week.

Paul,

if you could report back with your thoughts and any tips, that'd be great.

Will do Greg. The kids out of my hair for a little bit this weekend so I'll muck around with it and report back next week.

Had a play with it for about an hour or so I didn't get to explore too much or dare go into all of the options. I mostly centered my activity to the search engine and was mightily impressed at the amount of data that was at my disposal. Brian should be commended on his efforts. The fact that the info is time marked will make it a lot easier for researchers is my guess, not being one myself. The model idea is a good one in that it can be self managed for your own research purposes. I still find it a bit clunky but that might be more a criticism of my lack of understanding of the processes than anything else. As a student of this case I found it valuable for sourcing information. Far more specific than say perhaps a Google search.
I think the timeline is a great project. Anything that is designed to bring this case together like this does, is a good thing. I'll be sure to refer to it if I have to look something up.
Congrats again, Brian.

Paul McGurkenfarklein wrote:Had a play with it for about an hour or so I didn't get to explore too much or dare go into all of the options. I mostly centered my activity to the search engine and was mightily impressed at the amount of data that was at my disposal. Brian should be commended on his efforts. The fact that the info is time marked will make it a lot easier for researchers is my guess, not being one myself. The model idea is a good one in that it can be self managed for your own research purposes. I still find it a bit clunky but that might be more a criticism of my lack of understanding of the processes than anything else. As a student of this case I found it valuable for sourcing information. Far more specific than say perhaps a Google search.
I think the timeline is a great project. Anything that is designed to bring this case together like this does, is a good thing. I'll be sure to refer to it if I have to look something up.
Congrats again, Brian.

Thanks Paul.

My concern all along has been that there are many researchers like me who are less technically proficient than the average 10 year old. My sites are simple ones and run on a wing and a prayer. It can be the greatest invention since sliced bread, but not much good if those who need it most, have great difficulty using it.

That's not a criticism of Brian - just a sad fact about some of my generation. I agree he has done a tremendous job with getting it to work at all - and yep - it is an enormous amount of data that's been made available.

I hope I'm wrong. I hope it gets used and used to its best advantage. For myself, I just don't have the patience at the moment. I fear I would drive myself and everyone else nuts just trying to get my head around it.

greg parker wrote:

Paul McGurkenfarklein wrote:Had a play with it for about an hour or so I didn't get to explore too much or dare go into all of the options. I mostly centered my activity to the search engine and was mightily impressed at the amount of data that was at my disposal. Brian should be commended on his efforts. The fact that the info is time marked will make it a lot easier for researchers is my guess, not being one myself. The model idea is a good one in that it can be self managed for your own research purposes. I still find it a bit clunky but that might be more a criticism of my lack of understanding of the processes than anything else. As a student of this case I found it valuable for sourcing information. Far more specific than say perhaps a Google search.
I think the timeline is a great project. Anything that is designed to bring this case together like this does, is a good thing. I'll be sure to refer to it if I have to look something up.
Congrats again, Brian.

Thanks Paul.

My concern all along has been that there are many researchers like me who are less technically proficient than the average 10 year old. My sites are simple ones and run on a wing and a prayer. It can be the greatest invention since sliced bread, but not much good if those who need it most, have great difficulty using it.

That's not a criticism of Brian - just a sad fact about some of my generation. I agree he has done a tremendous job with getting it to work at all - and yep - it is an enormous amount of data that's been made available.

I hope I'm wrong. I hope it gets used and used to its best advantage. For myself, I just don't have the patience at the moment. I fear I would drive myself and everyone else nuts just trying to get my head around it.

I agree, Greg. I guess like most things it might become clearer (and easier) the more you use it and familiarise yourself with the processes. It will require some effort at first.
I am no researcher but I could see the benefits of the search facility as a keen student. However, as someone from your generation with very little patience nowadays, I was somewhat dumbfounded and confused by all the options. Something I take full responsibility for along with Father Time. He is a prick that Father Time. I wish he'd fuck off.

Hi all, thanks for the nice words, they're muchly appreciated.

Wow - well, the Hash Tags are kinda fluffy till we get back to the math.

Meanwhile, the essential capabilities are searching people by name, and searching events by time. Both of those things work. Everything works, it's just that the data isn't set up yet, and that'll still take a minute. (You can see it in progress though, I'll tell you about it in a microsecond).

The filter seems a little overkill at first, but you begin to appreciate it after a while. The thing that doesn't work yet is the Level filter (the 'zoom' capability), and that's because the data isn't set up yet. I'm trying to do things systematically, trying to cover the essential ground first. So far 2/3 of the Warren Commission is linked and in place, and the Motorcade Sequence is mostly done. (I discovered some noteworthy things about Costella's work, if anyone's interested). Right now we just reached 54,000 Hash Tags, so, that filter is going to be everyone's best friend.

Yes, the clumsy old mainframe-y look and feel... well, it allowed the site to get up and running quickly (which means, I get to focus on data entry instead of programming, and that's a good thing because the data is what needs work - the user interface can always be improved, we could easily do the fancy drag-and-drop stuff, but the big advantage of doing things this clumsy-looking way is that all the screens are cross-browser compatible. I've tried them on an iPhone, a BlackBerry, even a dumb old barely-access-the-internet 'antique' phone, and they all work. The only thing needed is Java (and your browser of choice) - I ran into a couple of issues with people using outdated versions of Java (for instance I was at a public wi-fi at the Whole Foods Market and they were back on version 5 or something, it made some of the buttons and fonts come out funny but other than that things still worked).

So now, I'm adding the '+' sign to the Search screens. This will blow your mind when it comes to the name fragments. (It makes for a real nice demo). In about a week, here's what you do: Let's say you have a name like 'Orcarberro' and you don't remember how to spell it, and barely remember how it sounds. Orca-something..... berry? Orcaberry? Like that?

If that's the situation, you go the Search By Name screen and type in Orca+ber, and also any of the following will work: Orco+Berra,Orc+Ber,Orka+Bero... you get the idea. The system will match to the database using 'LIKE' keywords on all of the tokens you type in (that are separated by + signs). And, it will take the intersection of the datasets and display the 'short list' for you (and if you wish you can then push a button to see the longer list).

The search options are only going to 'make sense' once you become familiar with the structure of the records. They're all alike, all very similar. Each object has three fields called create, update, and validate that hold the corresponding dates. In addition, each record has a latest audit id, latest archive id, and latest review id. The archive is where old audit records go, so they don't clutter the database (so far there are 170,000-ish audit records, about 144,000 of them are in Archive 1).

Let's say you're working in a 'busy' part of the database, like maybe people named 'Rodriguez' or events between noon and 1 on the 22nd. Normally if you go to one of the Manage screens and type in the first few letters of the name you get a reasonably short list, but for instance if you look for all events between 12:30 and 12:31 on the 22nd you'll get a pretty long list. SOME of those events pertain to the motorcade, and you can view those separately because of a naming convention, you can type 'MS' into any of the search screens and see "only" the records pertaining to the motorcade.

If you do this in the Search By Hash Tag screen, you'll see what's going on. Type 'MS' into the search field and hit enter. You're looking at an Event Bucket consisting of 64 events. That's what all that is. What it is, is mostly "cross references", so if you read the names you can see that many of those buckets end in 'People' or 'Events', these are the linked buckets that underlie the cross references.

Right now I'm about halfway through the MS motorcade sequence, but you can already see part of the real power of these advanced searches. FOR EXAMPLE - try exploding the MS Events bucket (you can get to it via the Events Guide, here: http://www.jfk-timeline.org/jfk_guide_events.php ). At first, on that screen, you're seeing the "data hierarchy only", which means in our terminology "attached buckets only". This is what allows you to surf up and down through hierarchical data structures, like the Warren Commission Report (which is in outline form, and you can surf through up and down through the outline layers beginning in the Evidence Guide, here: http://www.jfk-timeline.org/jfk_guide_evidence.php ). So NOW (we're still on the Explode screen), push the little button at the bottom that says "Show And Process Links" - and note how your display expands. You are now seeing (in addition to just the hierarchical data structure), the CROSS REFERENCES, in the form of People buckets and Event buckets and so on.

To see what this all means, you can now push the "other" button at the bottom, the one that says "Show Flattened View". Now, an additional table appears below the exploded Bucket, and it contains a list of all the People, Events, and Evidence "in" the entire data tree that begins with the 'MS Motorcade Sequence' event bucket. In other words, the system is GATHERING for you, everything that has anything to do with the Motorcade Sequence. And then it's showing you the list - here's the People, here's the Events, and here's the Evidence.

This is a "view" of the Motorcade, yes?

For a completely different view, use List Events or List Evidence, and set the alpha filter to 'MS' (with no other filter). If you do this on the Events, you get the 64 motorcade events in order, and the Evidence Buckets have pretty pictures too (that part is not yet completed, but you can see the first few) and so on. The List view reads more like a book, whereas the Exploded view is more for hard-core research, making sure all the i's are dotted and t's crossed.

The third useful thing (as mentioned earlier) is the 'Bucket Details' screen, which only allows you to look at one layer at a time, however it lets you surf up and down through the layers, so you can traverse a hierarchical data tree this way. (You can try this on the Warren Commission Report, it works to the extent the data's ready). When it comes to Events, this feature will be VERY useful - for instance, "the" Tippit Shooting is not just one event, but many, and if you want to zoom in on just one set of sub-events, this is the way you do it.

The data is organized as granular as is USEFUL. So, a lot of the fluff from Oswald's 201 file probably isn't going to be expanded much beyond where it already is. On the other hand, there are SEVERAL Warren Commission Exhibits that are just begging for further definition (I could rattle off a whole long list right off the top of my head, let's see there's the CE-5323 "Decker Exhibit" which consists of a bunch of vital affidavits from the Sheriff's Department taken the day of and day after and so on, there's CE-1024 and CE-2031 and in the Documents there's CD-5 and CD-7 and so on.... there's lots of "collections" that somehow ended up as individual exhibits - so what happens then is the data isn't granular enough, and if you look up the references for CE-1024 you get a whole long list of names, fifteen of them or something - and that shouldn't be).

So now, on that "flattened view" you just saw on the Explode Bucket screen, there's a little button that controls the duplicates, you can either show them or hide them. For instance let's say you have CE-1024. Every time that piece of Evidence appears in support of any of the Motorcade Events, the 'Explode' logic is going to recurse into the linked People Bucket and recover all those fifteen names every time CE-1024 appears. You can SEE all those duplicates, if you really want to know where they appear. You can count: this single piece of Evidence appears in support of 15 witness claims. That's not a good ratio!

So yeah, this is enabling some very hard core nuts-to-the-grindstone research, but if that's part of what you do then this is going to save you a WHOLE lot of time. Plus, as you say, just the Search features alone are useful, and I mean.... come on Greg, those Search screens only have one field on them! (Each - plus a button - what could be easier?) But yeah, if you want to see all the 'MS' records that were updated since last Tuesday 'cause you finally figured out you entered some bogus information but can't exactly remember where it was (or when)... you can find it. You CAN find it, with these screens. No more looking through file cabinets. Let your fingers do the walking. You'll see how this works when there are more cross references in place, you'll be able to do some pretty astounding stuff with "simple searches" (that are way beyond anything you can do directly on the evidence sites).

Anyway... miles to go... the Motorcade stuff should be done in a day or two, the Warren Commission's been lagging just 'cause there's so much of it (darn those guys, I hope they gag on a spoon - oh wait they're all dead already, oh well), and the next thing on my list is the DPD Archives, which has 40 boxes instead of 26 volumes (but there's less in a box than there is in a volume). This database is completely extensible, and I just wanted to drill down into the "most basic" areas, like the Warren Commission and the motorcade and that kind of thing. I really REALLY shouldn't be the one doing this, at some point we (the Research Community) are going to want the Subject Matter Experts to take over, 'cause I'm just a n00b and I dunno watta-da-fuggi'm-doin' when it comes to stuff like organizing the dataspace around George DeMohrenschildt. I didn't even want to do "this" much but I figured the Warren Commission is kind of like a hamburger, it's pretty hard to mess up once it's on the grill. So, I'll get the raw data records up there but I'm going to need help organizing them. The only way this resource is going to live up to its potential is if people decide it's a good thing and they have a reason to participate. I can provide "a" basic very simple reason, which is a search capability that's a bit beyond the evidence sites, but I mean, we could be thinking seriously out of the box with something like this, like... "how do we best use this tool in support of our efforts to re-open the Kennedy investigation and to get additional government documents released"? Well, on the latter part, I'm thinking, that this web site could be kind of like one of those landing strips you see on the rooftops around Christmas time, you know... "Santa come here, the arrows indicate the chimney". You know... ==> HERE <=== in big bright letters. Propaganda. Just like Lee Harvey Oswald.