Would it be too much to just count?

Garbage in, Garbage out

I guess it really doesn’t matter if we have redistricting reform or not if they’re going to be working with manipulated data.

15 thoughts on “Would it be too much to just count?

  1. It sounds less “manipulated” than scrambled.

    My take was that these algorithms were designed to protect against misuse by the bugaboos of Big Data.

    Yet, it sounds pretty obvious that the “cure” worked but the patient died.

    Maybe I am too complacent, but my emails are coming from places trying to sell me stuff I never imagined I would be interested in. And am still not.

    Annoying, infuriating but, for now, the cost of doing business in America. Other nations have put a lid on this and value individual privacy over business interests more than we do.

    So I guess this SNAFU is a botched extension of trying to protect us, although in the grand scheme of things, it won’t do much of that at all.

    Liked by 1 person

    1. If they were not collecting data in the census they have no business collecting, this would not be a problem.

      If they only collected the number of people, age and perhaps sex, then their would be no privacy problem. That’s all they really need for drawing districts.

      Maybe track race as well for the purpose of detecting Gerrymandering.

      But that’s all the Constitution really authorizes. It’s when you start collecting information about wealth and income, which have no place in drawing districts that there is data to exploit. And, of course, the modern census collects a lot more.

      Don’t collect the data and you don’t have to keep it private, and thus you don’t need to screw it up so it no longer serves it’s actual purpose.

      Like

        1. Well, why?

          If the redistricting is being done as it is supposed to be, by contiguous borders and common interests, then why should race be a factor?

          The information is only needed if you intend to redistrict by race, for one reason or another.

          Like

          1. Actually, I was agreeing with your, I thought, obvious point about keeping Gerrymandering from happening in the event of not properly applying an agreed upon formula for creating fair and balanced voting districts.

            Liked by 1 person

          2. “… by contiguous borders and common interests,”

            How do we determine common interests without additional data?

            Perhaps in rural areas where most are in agriculture and it’s support businesses. But what about metropolitan areas?

            Liked by 1 person

          3. @Jimmie and Len

            It seems to me you only need racial data to protect against racial discrimination if you have the data.

            My neighbor for 10 years was a homeowner, married, educated and had two well behaved and respectful children who were a joy to have as neighbors.

            So, whose interests were more in common, his and mine, or his and a single mother in a Richmond project, simply because he and she are black?

            Because the 4th Congressional district is drawn on the presumption that race matters more than everything we had in common.

            Like

          4. “Common interests”?

            I understand what you are saying. I grew up in Brooklyn, NY and except for a few years in Suffolk, VA. I have been in Norfolk: Ghent Square, Larchmont, Freemason, Ghent and now West Belvedere.

            Draw a line from any of these for a dozen or so blocks one way or the other and the demographic changes dramatically. Along with income levels, educational achievements, businesses, etc. Without all that info, a straight box would certainly not reflect common interests.

            The same is true in most major cities. This is particularly true in the last decades with gentrification putting upscale rents next to section 8 housing.

            Now, in a perfect world, it might be argued that the political lines should be drawn without consideration to such matters. So that there would be some poor, some rich, some whole families, some broken, etc. Then a politician would have to represent everyone’s interests in that district. A tougher job, but perhaps providing more equitable representation in the grand scheme.

            Yet you seem to be favoring common interests. And if that is the case, and I am not saying that is bad, how do we know that interests along 35th Street in Park Place are different from Mayflower Drive in Colonial Place without more details in the census?

            As always, the devil is in the details.

            Liked by 2 people

  2. The article says:

    “The problem, embedded in a computer algorithm, puts people where they don’t actually live and messes with age and racial demographics beyond an acceptable margin of error…”

    That’s not much to go on, but “algorithm” is probably the reporter’s word, or a word someone used with the reporter to try to explain what’s happening in layman’s terms.

    The way the description reads, I suspect the underlying data is fine, all recorded accurately and all nice and neatly organized for retrieval. Sounds like the problem is with the retrieval process. It is not uncommon for search queries (data requests) against an existing, perfect, database to be malformed. The resulting output can be garbled beyond comprehension sometimes, or can include inexplicable, random errors.

    The point is that the data collection process may be fine, but the data reduction (or processing) process may be flawed. Problems with the former would be cause for existential alarm. Problems with the latter, not so much.

    Put another way, the types and volume of census data being collected may be smart, but the engine for reporting it may be missing on a a cylinder or two.

    Liked by 2 people

    1. Sounds like an accurate description. But it also shows how tech is not always the best solution for simple tasks. Remember a few years ago, there were voting machines in VB that were CHANGING votes (for Scott Rigell, I believe).

      If the data retrieval is the issue, as you suggest, and makes sense, that should be an easy fix. But then again, I am not a tech guy, just an end user, like most of us,.

      Like

    2. Is the raw data undamaged so that new retrieval mechanisms can be developed?

      I think in terms of photography. Negatives in the old days and RAW files today. I can take digital RAW image and screw up a final image. But the RAW is unaltered and will always stay the same. So I can try another Photoshop effort or technique.

      Liked by 1 person

        1. RE: “Maybe new AND better more accurate?”

          Not sure what you’re getting at. There are two processes involved in the puzzle: 1) collecting and recording the census data, and 2) retrieving it in different ways. The article suggests the problem is in the retrieval process.

          The reason I commented was to point out that the technology is not necessarily failing us. It’s not like the problem is too hard, or that the system design is too complicated or tries to do too much. There doesn’t appear to be any strong reason to rethink the original puzzle or to create a different solution. More likely, the errors are the result of simple typos, such as missing “punctuation” in lines of code.

          I only wanted to convey that the problem the article describes doesn’t appear to be of a type that requires throwing the baby out with the bathwater.

          Like

          1. Sorry, I wasn’t clear. I was referring to the retrieval.

            Yes, do not thrown out the baby with the bathwater. But fix the retrieval issue. Now that it has been identified, I would expect the fix to be relatively easy for those who do it for a living.

            Just trying to be optimistic in pessimistic times.

            Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s