The first rule of good software design? First of all, do no harm!


Who among us hasn’t been cursed at some point by the self-correcting gremlin?

SMS message fails

You hit a “perfect letter” message, thumb your thumb on “SEND”, then watch in slow motion the auto-correcting gremlin that decides you couldn’t mean THAT, right? A moment later, he replaced your words with something ridiculous before sending it on his way before you could object.

It’s like having a little Mr. Bean in your pocket.

I especially like the following example – mainly because this has not happened to me.

This is one of those cases of product designers trying to help us, but accidentally doing the opposite. A misspelled message of ‘I’m leaving no‘would have been as careless, but not as heart attack inducing as’ I’m leaving you’. Less funny too.

Still, as funny as AutoCorrect failure may be, I suspect that the full impact of AutoCorrect is a victory. The advantage of fixes outweighs the failures.

Sometimes these attempts to help can be more damaging.

So, is it a date?

Formatting dates

One of the new “user interface wizards” we’ve seen over the past decade is the ability for applications to recognize dates in plain text.
Although as humans we switch between date formats with relative ease, it is only recently that software has become sophisticated enough to understand that the following all refer to the same time:

  • Wed Jan 25
  • 25-1-2017
  • Wednesday, 25/01/2017
  • 1/25

In fact, apps like Gmail, MS Word, and Slack are even smart enough to understand concepts like “tomorrow” or “next Friday” as dates you can click to turn them into events or dates.

It can be very useful – until it is not …

Please stop helping

Although Microsoft is no longer quite the global juggernaut it once was, its ability to offer its software at a very affordable price to schools, universities and universities has kept them dominant in education and research. . The vast majority of the world’s academic studies and articles are produced using a combination of MS-Word, Excel, or Access. A feather for MS caps.

Unfortunately, last year it was reported that Microsoft’s Office suite was responsible for breaking or invalidating around 20% of genetic studies since 2004.

How did it happen?

As with most sciences, genetics has its own language and vocabulary of terms that most of us will never come across. For example, you probably didn’t know …


SEPTIN2 looks like a party mask.

Septin 2 (or SEPT2)

  • is a protein that appears to be used in the tracking and analysis of tumors and cancers (yes, I’m irrelevant here). It is commonly referred to as “SEPT2” in studies. Looks like a little masquerade mask.
  • Annular associated with the membrane (C3HC4) 8 (or MARCH8)

  • is a gene found in creatures, from mice to trout to humans. It may have antiviral properties and is often, for good reason, abbreviated as “MARCH8”.
  • Many of these studies are tens or even hundreds of pages long, and individual Excel tables may list “SEPT2”, “MARCH5”, or “MARCH8” several thousand times.

    How does Excel handle these gene names?

    As you have probably guessed, Excel and MS Word attempt to convert each of these genes into a useful date. In fact, in total, there are at least 25 different known gene names that are automatically renamed when imported into Excel.

    “It’s not a gene, sir – it’s a calendar event !!”

    The real kicker? When Excel automatically converts the gene to a date, it permanently forgets the original state. You can convert it to 25 different formats from September 2, but “UNDO” just removes the text rather than returning it to the original “SEPT2”.

    Of course, if you know how to turn off the defaults before importing your data, you won’t have any problems. Many researchers know how to do this. Yet this problem has been detected in 720 of the 3,600 genetics articles since 2004, so it is not an “abnormal event”.

    We will never know precisely what impact this has had on genetics and medical research. But corrupted data MUST just weaken the quality of the results. Scientists rely on data from other scientists, and we know the data is wrong.

    Would we be any closer to a cure for cancer or antibiotic resistant superbugs if all this corrupted data were clean?
    Probably – although we never know exactly how much.

    “First of all, do no harm”

    This idea is one of the central principles of medicine. In other words:

    “Better to do nothing than do something that might make it worse. “

    We don’t have a set of universal principles that we adhere to in software design, but damn it… that’s a good first design principle, isn’t it?

    “When designing new features, do no harm first.”

    Source link


    Comments are closed.