Everyday Scientist

18 Comments »

RSS feed for comments on this post. TrackBack URI

1. to say “deconvolute” instead of “deconvolve” in the context of Fourier analysis. Deconvolute sounds awful, like you’re made the problem worse and now you’re trying to go back and fix it. Sam Lord discusses the deconvolute vs deconvolve debate in this post

   Pingback by Quantum mechanic — June 7, 2007 #

2. Sam at Everyday Scientist blog writes:

   Pingback by Nautilus — November 30, -0001 #

3. Why not just “separate” the fluorescence lifetime and instrument-response function?

   db

   Comment by David Bradley — June 6, 2007 #

4. I looked through some books on signal processing in chemistry. It seems like they mostly go out of their way to avoid using either term, finding ways to use deconvolution or convolution. When they do use one, however, they use convolute.

   Convolute does not necessarily mean to make more complex. It literally means the same as convolve – to roll together.

   Comment by Colst — June 7, 2007 #

5. Colst: Yeah, I know that they both denote “roll together,” but—at least to me—they have different connotations. “Convolute” has less of a mathy/sciency meaning … a more everyday usage.
   .
   David Bradley: How are we supposed to separate ourselves from the masses unless we use overly complicated words and jargon?? ;)

   Comment by sam — June 7, 2007 #

6. I’m with deconvolve!

   Comment by PhilipJ — June 8, 2007 #

7. […] Sam at Everyday Scientist blog writes: We were discussing some grammar at Chemical Physics Journal Club this week: which is the (more) correct sentence? 1. It is important to deconvolute the fluorescence lifetime from the instrument-response function. 2. It is important to deconvolve the fluorescence lifetime from the instrument-response function. I think sentence 2 is better. To me, “convolve” is to (usu. mathematically) roll together multiple things, while “convolute” means to make complex: you can convolve two mathematical functions or signals, and you can convolute a sentence. (Unfortunately, the noun form of each is “convolution.”) […]

   Pingback by Nautilus: Convolute or convolve? — June 8, 2007 #

8. I’ve always used deconvolution (not so much deconvolute). It’s the word that everyone around me was using and a book by the same name is where I picked up many of the routines I used on spectra.

   Comment by Deepak — June 8, 2007 #

9. My four cents goes with deconvolve, for the way it rolls off your tongue. That or separate (they don’t put mechanically deconvolved chicken in my Vienna sausages).

   Comment by kendall — June 8, 2007 #

10. I try to skirt saying it, because most of the other chemists I work with aren’t physically oriented enough to care. If it comes up, I point out that something is ‘convolved with’ something else, or that a quantity is the the result of the “convolution of x and y”, and probably would say that one has to “deconvolve” them, or “take the deconvolution”. I wouldn’t say “deconvolute” because it sounds funny to me.

    I have found signal analysis techniques really handy as a chemist. The sort of mental framework they provide is broadly applicable to spectroscopy, electronics, mechanical systems, etc. I encountered them in EE courses, though. I’d recommend a circuit analysis sequence to science majors able to tolerate the math, just to get at the transform methods, if your chem department doesn’t explicitly teach them.

    Comment by Dave Eaton — June 16, 2007 #

11. very interesting, but I don’t agree with you
    Idetrorce

    Comment by Idetrorce — December 15, 2007 #

12. I agree with you, as do the majority of web pages in the image processing field. A Google search for “deconvolve the image” finds 3460 hits, while “deconvolute the image” finds only 22. Thus, in the popular vote, you are ahead by over 100:1.

    Comment by Hankk — January 6, 2008 #

13. hey, thanks for helping settle my lab’s dispute over this useage. it’s right up there with “novel” and “facile” for my least favourite science buzzword.

    Comment by Joel — March 19, 2008 #

14. My friends, the answer is so obvious – just look at how verbs are formed from other nouns ending in “-ution” in the English language.

    The verb form of revolution is “to revolve,” not to “revolute!”

    Why then would deconvolution break the trend?

    Comment by Elizabeth — July 11, 2008 #

15. In my mind, deconvolve is a process (or more reflexive in nature), and deconvolute is an action; to clarify that a bit more, I would argue that:
    if you deconvolute a bunch of functions,
    said functions deconvolve.

    I don’t have any hard argument for that, just a gut feeling (and English is not my native language; but then, I hat Latin longer in school than English ;-) )

    -j.

    Comment by Jens — November 2, 2008 #

16. Deconvolution has a specific mathematical definition. Deconvolution in scientific papers should not be interpreted colloquially. Scientists who process data understand it differently than the guy on the street.

    Separate also has a specific chemical definition.

    Once you gain the vocabulary and experience, it will make more sense and people who use the vernacular definition of scientific words will make you laugh.

    Comment by DR — January 16, 2009 #

17. This is absolutely not true. By studying the history of language, you would know that words are often bent from their original usage to convey ideas. Colloquial or not, “deconvolve” is a great word and you can’t claim it exclusively for mathematics. Many common words in our language today were recast from their original meaning.

    I think many of the great writers of English literature would not support your hard line on word usage. If you can use a single word to replace sentences that describe an idea, by all means….do it! Laugh if you want to, I really don’t care.

    Comment by SH — February 14, 2010 #

18. Frost = to make frozen
    Defrost = undo the frozen making

    Convolute = make more confusing
    Deconvolute = undo the confuse making

    Convolve = roll things together
    Deconvolve = undo the rolling of things

    I prefer to make less confusing, because that what I hope will happen deconvoluting the data!

    Comment by breze — November 5, 2011 #

Leave a comment

Subscribe: