Slashdot Mirror
Arlington National Cemetery's Many IT Flaws
imac.usr writes "A story in today's Washington Post calls to light the utter failure of the nation's most sacred final resting place to modernize its pen-and-paper record system. According to the story, the cemetery's administrators have spent $5 million without managing to accomplish the seemingly simple task of creating a database record of the site's graves. As Virginia senator Mark Warner points out, 'We are one fire, or one flood, or one spilled Starbucks coffee away from some of those records being lost or spoiled.'"
The government gets huge economies of scale. That's why we should have them in charge of the health care system. Clearly we will be able to save substantially more money than the private sector once the profit motive has been removed.
For all of our soldiers who have earned the right to be buried there and we can't even get a decent IT system in place to help people or keep such important records.
To our Fallen Hero's.... I am sorry.
I do think it is time that companies and even people stop being so damn greedy and do their jobs. Granted we may not have the insight as to what is happening directly but I am left to wonder who is asleep at the controls on this one. We have private sector people doing jobs that are comparable size to this job and I am sure 5 million dollars would have paid for their time and a mojito and Starbucks coffee whenever they wanted it. I think it is time to disband our Government and reform with people that a hell of a lot more honest then some of the guys we have in there now. Sorry to make this political but the fact remains that someone is not doing their job. Any person's loved ones are important to them but a person who defended our rights and country (regardless if the war is right or wrong to which those that feel it is wrong it is time to bitch at the civilian leaders case and point would be the recent Gen. McCrystal deal.) and we can't honor them with keeping accurate records and spending money WISELY when it comes to their final resting place. Sad...
And there are already systems available that can manage cemeteries so why not purchase one?
If builders built buildings the way programmers wrote programs, then the first woodpecker would destroy civilization.
Because its government owned and operated and the US government is an utter joke when it comes to financial decisions. We bail banks out so they can throw crazy huge parties with the celebrities while our citizens lose their homes because they cant afford to pay last year's taxes.
You see a hobo on the street corner begging for money every day, claiming that he's down on his luck and can't seem to do anything to change it.
You buy him a brand new house, and GIVE it to him, free and clear, to end his days as a transient.
He then sells the house, and blows all the money on drugs, booze, smokes, and hookers.
You then see him back on the same street corner, begging for money, claiming that he's down on his luck and can't seem to do anything to change it.
The "surprise" ending is that the hobo is the U.S. government, and the money is our taxes. They need to raise them? Like hell they do.
And yes, it was spent on drugs, booze, smokes, and hookers.
It might not be within the strictest interpretation of their (NA's) charter, but I think its certainly within the spirit of their mission.
THIS does not happen enough in the Federal Government. 95% of the time when an agency is in need of a skillset that is outside its purview (or sometimes within its purview, but present in a different department), it contracts it out to some third party vendor with questionable skills and typically high prices. Every federal agency should be ready to consult for other agencies when its primary skills are in need, but it almost never happens that way.
Where's accountability when 5 million gets spent and nobody can even make something as simple as a SPREADSHEET?
Clearly you don't work in or understand IT. First there have to be meetings. Lots and lots of meetings. First at management level to initiate the project. Then detailed meetings to set up staffing and outline goals. Then middle management needs to be appointed (more meetings) so that they can flesh out those goals in more detail (more meetings). Of course this is after HR recruits the middle management. The middle management goes through the same process to recruit actual staff. Then management meets with staff that provide feedback on those tasks "No I'm sorry you can't magically walk around with a laptop and scanner and have it absorb names off the gravestones. No there's no technology to do that on the horizon". Then middle management needs to report back to senior management (did I mention meetings?) and senior management needs to meet separately to decide what it means to the project. At this point all those discussions will get confusing so will need to be summarised and corrected. Only now can we start to see a plan coming into being (drafted by middle management, approved by senior management. You guessed it more meetings). At this point work may commence but if it is it will typically be halted by a new priority/requirement being pulled out of senior management's rectum^H^H^H^H^H^H^H^H^H^H^H ahem I mean coming to light. This will totally screw up every agreement made about the direction and even nature of the work, which will require more meetings at all levels to sort out.
Oh and don't be fooled this happens in industry as well as government. Privitising just adds another layer to all this mess and provides another opportunity for waste each time someone changes their mind or adds an unreasonable or ill thought through requirement.
$5 million is nothing. The fact that an intelligent 6th grader could do better is by the by. it's not how the world works.
These posts express my own personal views, not those of my employer
A spreadsheet? Seriously? WTF, you're just one column sort away from losing data integrity. It's how every Excel file dies.
'We are one fire, or one flood, or one spilled Starbucks coffee away from some of those records being lost or spoiled.'
This is not an IT problem. This is a basic information storage problem dealing with backup procedure. If you're a major organization and you don't have copies of your records, whether paper copies, microfiche copies (which seems to be the case here), or electronic ones, you're vulnerable.
Similarly, IT doesn't necessarily solve this problem. If you digitize all the records to a single server and don't make proper backups, you could still be one fire or flood (or even a coffee) away from losing the records.
(Btw, I do realize that original paper records may have some value as historical artifacts themselves. But those should be in an archive somewhere protected from floods, fires, and errant cups of coffee, while people accessing these records on a daily basis should be using copies, whether digital or microfiche or whatever.)
The jackasses were moving existing remains, and putting new remains into old graves, and putting landscaping on top of existing graves.
When your problem is that the guy with the backhoe hasn't been trained to speak up if there's already an existing grave where he's been told to dig a new grave, the solution isn't "hire some geeks to program some computers".
The solution is to create an environment where everyone who works there respects the dead, and to make sure that procedures enable that respect. It's not an IT problem at all.
Not to downplay historical significance, but it has no relevance to databases. I've managed multi-millions of records for a small company using MSSQL 2000, and with 200+columns in some of those tables. 400k records should be quite trivial from a technology standpoint. A single database should be capable of storing the data for the entire national cemetery system.
It's more a matter of determining what needs to be stored in the database, which is a planning question rather than a tech question. If the off-the-shelf packages are unable to scale to that size, they were designed not to scale (rather than not designed to scale).
As another has pointed out - a few hundred k records is a trivial problem for a database. You could probably do this in MS Access (though I wouldn't recommend it).
It isn't like Arlington has to deal with 10k bodies drifting in and out of the cemetery every hour or something like that. This is just a big table indexed for easy searching by name/location, and it gets a couple of inserts per day.
If this takes more than a day or two to put together, somebody is doing something wrong.
Good gravy man, that's like table creation 101. A few hundred thousand records with mostly read operations, rare updates, occasional inserts? That problem has been solved, thousands of times already.
Just grab some digital collections software (Omeka) or customize some Drupal or code up a quick model in Rails/Django and you're in business. Omeka would probably be the quickest startup and provide the capability for tons of rich metadata.
No want/need for a public web interface? Don't set one up. Done. Lunch time!
Except for the data entry, that's going to be some fun data entry.