College Physics Instructor and Sewing enthusiast.

20 thoughts on “About”

  1. Very interesting site, wonder why I’ve not found it til now? I’d love to see (and possibly try) your shirt pattern; is it available? As Michael P says, I’ve done many similar experiments on collar and stand shape (none as extreme as yours) and shared my results in a new book; would be happy to compare notes:)


    1. Thanks for inquiring. As with any project, there is always one more improvement. My latest tweak has to do with the sleeve placket. I was looking at how they seem to be done on RTW shirts. At first the technique seemed simple enough. However, after several trials, I realized that I don’t like it.

      The other factor holding me up is that I feel I need to develop at least a minimal set of instructions to go with the pattern. Especially since my pattern is a bit different from traditional sewing patterns.

      Let me take this opportunity to explain what I am trying to do. I have created a program that will create a shirt pattern to a specified set of dimensions. Essentially a custom pattern. At this time the only source of body dimensions I have access to are the ASTM sizes. This gave me a way to “test” the pattern algorithms over a range from 34 to 52 although not actually test the pattern on a real person. My goal is to eventually be able to offer custom patterns. That goal is probably a year (or more) away.

      What I do have available are the 37 patterns in the ASTM sizes from 34 to 52 in half inch increments. My thought was, that by suitable choice of finished garment dimensions, that one could get a well-fitting set of basic blocks for the shirt (yoke, front and back). It seems to me that the “Back shoulder width” is an important dimension to get right. So instead of choosing the pattern by the chest size, identify the pattern with the correct “Back shoulder width” in the “finished garment dimensions table”. That way the shirt will fit well over the shoulders. Other adjustments may need to be made.

      I have been procrastinating on making the patterns available as I tend to feel that there is always one more tweak. I should probably have the patterns done in the next few days.


    1. Yes, that is the intention. I know that it works for me. However, the other patterns have been drafted based on statistical body sizes. I believe that it should be the same. Alas, I have not tried it out on any other real person.

  2. An interesting choice, to be sure:) I’ve played around with something somewhat similar, but it seems to me that any curve-difference between the back’s top edge and the yoke’s lower edge, if it’s to fit well, could only be determined from a carefully customized shoulder fit, since some back shapes don’t need any curve at that seam, and others need plenty. Plus, it’s such a non-typical look and shape that the pattern as a whole will be less likely to be regarded as a “basic” shirt, by most people, I’d guess. For me it ADDS interest, but I’m probably not a typical shirt pattern user.

    1. Actually I agree. This pattern may be just a quark and of little interest in general. However, the problem of the yoke shape is what got interested in sewing and design. Then, once I got into the design process, my attention turned to producing a wide range of pattern sizes. This being the only pattern I had in hand, was the one I used. The whole process has been an interesting study with so much to learn.

      Thanks again for your interest.

  3. I hear ya! Still, after all this effort, in your shoes, I’d include a standard yoke shape as well, so the whole thing can be regarded as a basic and classic shirt pattern graded across a wide range of sizes; a very useful thing to have available.

  4. I can open with the same line as David: Very interesting, and I’m kicking myself for not finding this site sooner.

    I am currently designing a shirt pattern myself. Not that I really know what I’m doing, but I find it’s a nice way to learn :)

    Your experiments are very interesting, please keep them coming! And thank you for making this available for the rest of us.

  5. I kind of forgot to wade into the whole “I have created a program that will create a shirt pattern” thing.

    I am doing something similar. I have a website that creates patterns, based on measurements that you can enter.
    It is not a commercial endeavour. I am, much like you, simply trying to make more options available to (menswear) sewists.

    Your scientific approach is fantastic. I love all the work you’re doing, and if you see room for synergy, I’d love to hear your thoughts.

  6. Thanks for taking a look. Like you, this is a part-time avocation for me so things don’t happen fast. I would love to find a way share the ideas and experiences. One of the things I find daunting for something like a shirt pattern is myriad of design choices for the parts like collars and cuffs. The basic blocks of the shirt were not hard. The sleeves are a bit tricky, getting the sleeve – arm scye right and all. But then one has to make choices for the collars and cuffs and make all the pieces fit together. It is all part of the challenge.
    I am ready to take break from sewing shirts for a while and I really need a pair of trousers so I think I will give your “Textbook Trousers” pattern a try.


    1. Hi, Martin. You said: “But then one has to make choices for the collars and cuffs and make all the pieces fit together. ” Forgive the plug, but that’s exactly what my new book is for:) It’s a collection of detail options, as you describe, complete with full-size patterns to print out plus extensive shape analysis and exhaustive construction and “making all the pieces fit together” info, all of which assumes one already has “The basic blocks of the shirt…” I hope in the near future to test out your pattern (and Joost’s, to be sure!) and to report my experiences in my blog about them both.

      1. It seems to me that the art of shirt design is in the collar and cuffs as these are what gets noticed first. The development of the yoke-front-back and sleeves is more engineering than art. What I was trying to get at in my reply to Joost is that developing an algorithm to make the yoke-front-back is relatively straight forward as there are few measurements and few design choices (for a classic dress shirt). The algorithm that matches the sleeve to the arm-scye is more challenging. After that come the neckband-collar and cuffs. Here is where design choices must get made that come down to more art than engineering and this is where I am least confident and most uncomfortable about committing to any choice. Hence, as I develop the algorithm I try to build in the greatest possible flexibility for future changes. I find planning for unknown future options to be one of the most challenging aspects in the design of an algorithm that produces a shirt pattern.

        What I would like to do is produce a pattern that just consisted of the yoke-front-back and sleeves and leave the sewist to choose/develop the design of the neckband- collar and cuffs following the suggestions of your book. As I refine the program I hope to find ways of both communicating this freedom to the sewist and enabling the sewist to effectively explore these design possibilities. However, I did produce a shirt pattern and, for a pattern to be complete it must come with some options for collar and cuffs. Thus the pattern that I produced has collar and cuff options that work although I do not like having to commit to such choices. I would much rather leave that to the sewist to explore.

        1. I would like to make the program available to the public and I have been considering how to do that. I won’t release the script in its current form for several reasons:

          1) My script is ugly. It works fine but it is not pretty or user friendly.

          2) Occasionally it crashes illustrator. It seems to be pushing the bounds of illustrator will do, at least on my six year old machine.

          3) The domain of the input parameters needs to be identified. I would not want a user to input parameters that cause the script to produce garbage. However, the script has about 15 input parameters with interdependencies so the problem of identifying the domain does not seem simple to me.

          I would like to do something like Joost and have an interactive website where a user could upload their parameters and have the pattern produced on the spot. To do that I would need to redo the script in SVG. That project will probably need to wait for summer.

          The closest I can come to that in the near future will be to take requests for custom patterns which I will do once I put up the post explaining the needed measurements. That should happen sometime in the next few weeks.

          1. Very cool; may I ask to be on your list as a potential beta tester? (of both custom patterns and your script reworking; I run Adobe CC on a relatively new iMac)

  7. I have the same problem, my code is horrible and i don’t want to share with anyone. ;-)

    Do you know this project: http://www.taumeta.org ?
    TauMetaTau uses Inkscape. I did some tests extending the original code because it’s open source and i like it.

    I recomend you to use some program language that is more flexible so you can achieve your plans.

    Thank you

Leave a Reply

Your email address will not be published. Required fields are marked *