Twice this week, I have come across embarassingly bad data (successfulsoftware.net)

by hermitcrab 64 comments 84 points
Read article View on HN

64 comments

[−] stared 47d ago
I dislike the premise. I mean, good data is wonderful.

But if institutions are expected to release clear data or nothing, almost always it is the later.

What is important, is to offer as much methodology and caveats as possible, even if in an informal way. Because there is a difference between "data covers 72% of companies registered in..." vs expecting that data is full and authoritative, whereas it is missing.

(Source: 10 years ago I worked a lot with official data. All data requires cleaning.)

[−] Mordisquitos 47d ago
But surely we should expect some basic sanity checks on published data? This isn't some petrol stations being placed in the middle of a field due to minor typos or bad rounding, or some petrol stations' prices being listed as all 1.00 £/l out of laziness, or even a case of all unknown locations being listed as 0°0'0" N, 0°0'0" E by default. What the author reports appear to be mistakes which should be rather trivially detectable on input.
[−] ZiiS 47d ago
The problem is stats can actually do more with all the data including obvious errors. If you start filtering out data where they miss entered lat log you might introduce a new bias.
[−] chaps 47d ago
Sure we should indeed expect that they do that. But look at enough data and you'll learn that those expectations are a path towards never-ending frustration. I've been there, spending >100 hours cleaning data... that never got published because I was too damn focused on the dozens of years of errors that many, many people created.

To be clear, I'm not saying that we should accept messy data. Just, reality is messy and it's naive to think we can catch and remove all of reality's messiness -- which includes the bureaucratic slop that led to the data being published in the first place.

[−] freehorse 47d ago
I don't think these issues are close to the issues the article talks about. The author does not talk about data coverage, data collection methodologies or missing values or whatever, but data that is actually wrong, ie location coordinates, prices, numbers that make no sense. Including swapping latitude/longitude and wrong decimal points in numbers.

On the other hand, I agree that bad (but usually fixable) data is better than no data.

[−] stared 47d ago
Yep, expect in real data actually confusing columns, NaNs casted to values like 1673, duplicates, etc, etc.

I prefer to get data with swapped lat/lng (a trivial fix), or prices said in dollars but being in cents, to no data.

[−] readthenotes1 47d ago
I read the premises as "1. at least look at it 2. Have a way to fix it"

Those seem reasonable asks.

Edit to add: the tragedy of the school in Minab is an example of how bad things can go--and it just hints at how much worse bad data can bem

[−] stared 47d ago
For me, it sounds much more like "vibe-bombing" or in general a lot of pressure to select targets, without necessary time for due diligence.

Vide https://news.ycombinator.com/item?id=47544980

[−] readthenotes1 46d ago
My understanding is that the school was a part of the military base over a decade ago and was therefore put on a target list.

The list was never updated when the building was turned into a school.

It wasn't vibe bombing, and there certainly was enough time to do due diligence, but there was no process in place to do so.

[−] sd9 47d ago
Agreed, pretty much all data is flawed. I still want my hands on it.
[−] chaps 47d ago
I have mixed feelings about this. On one hand, yeah stop publishing garbage data, but as a FOIA nerd... I'll take the data in any state it is. I'm not personally going to be able to clean the data before I receive it. Does that mean I shouldn't release the unsanitized (public) data knowing that it has garbage data within? Hell no. Instead, we should learn and cultivate techniques to work with shit data. Should I attempt to clean it? Sure. But it becomes a liability problem very, very quickly.
[−] hermitcrab 47d ago
So you expect the 1000s of people trying to use the fuel price data to each individually clean and validate it, rather than the supplier doing it?
[−] yorwba 47d ago
One of those people can republish their cleaned and validated version and the 999 others can compare it to the original to decide whether they agree with the way it was cleaned or not.
[−] chaps 47d ago
What...?
[−] torginus 47d ago
What does it mean to clean the data?

Do you remove those weird implausible outliers? They're probably garbage, but are they? Where do you draw the line?

If you've established the assumption that the data collection can go wrong, how do you know the points which look reasonable are actually accurate?

Working with data like this has unknown error bars, and I've had weird shit happen where I fixed the tracing pipeline, and the metrics people complained that they corrected for the errors downstream, and now due to those corrections, the whole thing looked out of shape.

[−] GMoromisato 47d ago
Clean data is expensive--as in, it takes real human labor to obtain clean data.

One problem is that you can't just focus on outliers. Whatever pattern-matching you use to spot outliers will end up introducing a bias in the data. You need to check all the data, not just the data that "looks wrong". And that's expensive.

In clinical drug trials, we have the concept of SDV--Source Data Verification. Someone checks every data point against the official source record, usually a medical chart. We track the % of data points that have been verified. For important data (e.g., Adverse Events), the goal is to get SDV to 100%.

As you can imagine, this is expensive.

Will LLMs help to make this cheaper? I don't know, but if we can give this tedious, detail-oriented work to a machine, I would love it.

[−] torginus 47d ago
Data and metrics is 90% what upper management sees of your project. You might not care about it, and treat it as an afterthought, but it's almost the most important thing about it organizationally.

People who don't heed this advice get to discover it for themselves (I sure did)

IF you can't make the data convincing, you'll lose all trust, and nobody will do business with you.

[−] Phlogistique 47d ago
That it's it's better to publish the garbage data than to not publish it though. I would worry about complaining too much lest they just decide to stop publishing it because it creates bad PR.
[−] albert_e 47d ago
Concluding passage:

> Authors should have their work proof read

Agreed.

Opening passage:

> A quick plot of the latitude and longitude shows some clear outliners

"outliners"

Ouch!

[−] bobro 47d ago
This article assumes that there is a person with dedicated time to validate the data. Imagine you want this data and ask for it, but the government says, “sorry, we have this data, but we read an article that said we can only publish it if we spend a lot of time validating it. This data changes frequently and we don’t have a chunk of a full-time data analyst’s salary to spend on it, so we just aren’t going to publish anything. We’d rather put out nothing than embarrass ourselves, so you can’t even try to validate it yourself.”
[−] mlaretallack 47d ago
I saw the RAC one this morning, though I was miss reading the graph, as why would the RAC publish such an obvious mistake.

I have written my own Home Assistant custom component for the UK fuel finder data, and yes, the data really is that bad.

[−] bobosola 47d ago
A couple of days after the UK Fuel Finder service launch last month, I wrote a hobby site using its API to get the cheapest local fuel prices: https://fuelseeker.net. I too discovered prices which had obviously been entered in pounds rather than pennies, or even missing altogether some cases. You would think that they could have done a bit more basic data cleansing on the server to catch that type of thing.

But, hey, we’re all wise after the event. To their credit though, they do seem to be actively reacting to feedback. I also contacted them about the bad data issue, and they are now adding user warnings about bad price values at the point of data entry (according to https://www.developer.fuel-finder.service.gov.uk/release-not...).

[−] hermitcrab 47d ago
Why did the title of this post get moderated from:

"Stop Publishing Garbage Data, It’s Embarrassing"

To the rather lamer:

"Twice this week, I have come across embarassingly bad data"

?

[−] alias_neo 47d ago
I was looking at that RAC chart this morning. Given it's Sunday, and I was reading before my morning coffee, I'm not ashamed to say it took me a good few seconds of zooming in and out to realise they'd used a decimal point where a comma should have been.

Easy type to make, but seriously, does no one even take a cursory look at the charts when publishing articles like this? The chart looks _obviously_ wrong, so imagine how many are only slightly wrong and are missed.

The fuel prices one could surely be solved with a tiny bit of validation; are the coordinates even within a reasonable range? Fortunately, in the UK, it's really easy to tell which is latitude and which is longitude due to one of them being within a digit or two of zero on either side.

[−] Frank-Landry 47d ago
Did a bot write this title?
[−] hermitcrab 47d ago
If you are putting out data without doing even the most basic validation, then you should be ashamed.
[−] ath3nd 47d ago
[dead]