The Challenges of Presenting a Pattern
I think every knitter at some point in their lives has been frustrated by a pattern they couldn't understand. Perhaps the wording was vague and a chart would have helped, or maybe the chart was unclear and written instructions would have been better. Maybe the pattern designer assumed familiarity with a certain technique when a more thorough explanation would have helped.
Though many of these challenges are surmountable, some of them are not. What if the pattern were written in a language completely foreign to you, and translation software weren't much help? What if you were visually impaired and your voice browser couldn't read charts, and no written instructions were provided? What if you simply could not figure out what the designer meant by some terminology, and you couldn't get a hold of the designer? You either would have to go to extraordinary measures to get the pattern to work for you, or, more likely, you would give up and move on to a different pattern.
It's unreasonable for us to expect that designers can write a pattern to fit every knitter's preferences. Even if a designer managed to publish a pattern with written instructions in 5 different languages, a chart, English and metric measurements, summary instructions with detailed explanations, all in 5 different sizes, there may still be factors that she didn't consider.
Consider combination knitting, for instance. Knitters who knit combined tend to prefer that decreases are expressed in terms of their lean rather than the technique itself. For instance, a k2tog to non-combination knitters implies a right-leaning decrease. Under certain circumstances in combination knitting, a k2tog, if taken literally as a technique, will actually produce a left-leaning decrease. One solution would be for a right-leaning k2tog to be expressed in the pattern as k2tog-R. While this may be an optimal notation for a combination knitter, it may be more confusing for a Western knitter unfamiliar with this more uncommon convention. What is optimal for one person is not optimal for another.
Considering the wide range of people's preferences, wouldn't it be great if everyone could use a pattern in the format they wanted most? Wouldn't it be incredible if the knitters themselves were empowered to do this, rather than putting the burden on designers and publishers? Is this too much to ask? Is this impossible?
No, it's not. KnitML can and will solve this problem.
Though many of these challenges are surmountable, some of them are not. What if the pattern were written in a language completely foreign to you, and translation software weren't much help? What if you were visually impaired and your voice browser couldn't read charts, and no written instructions were provided? What if you simply could not figure out what the designer meant by some terminology, and you couldn't get a hold of the designer? You either would have to go to extraordinary measures to get the pattern to work for you, or, more likely, you would give up and move on to a different pattern.
It's unreasonable for us to expect that designers can write a pattern to fit every knitter's preferences. Even if a designer managed to publish a pattern with written instructions in 5 different languages, a chart, English and metric measurements, summary instructions with detailed explanations, all in 5 different sizes, there may still be factors that she didn't consider.
Consider combination knitting, for instance. Knitters who knit combined tend to prefer that decreases are expressed in terms of their lean rather than the technique itself. For instance, a k2tog to non-combination knitters implies a right-leaning decrease. Under certain circumstances in combination knitting, a k2tog, if taken literally as a technique, will actually produce a left-leaning decrease. One solution would be for a right-leaning k2tog to be expressed in the pattern as k2tog-R. While this may be an optimal notation for a combination knitter, it may be more confusing for a Western knitter unfamiliar with this more uncommon convention. What is optimal for one person is not optimal for another.
Considering the wide range of people's preferences, wouldn't it be great if everyone could use a pattern in the format they wanted most? Wouldn't it be incredible if the knitters themselves were empowered to do this, rather than putting the burden on designers and publishers? Is this too much to ask? Is this impossible?
No, it's not. KnitML can and will solve this problem.
Comments
I really love your enthusiasm - your blog post reads like a knitting manifesto :)
Actually I have been following your KnitML efforts from the beginning and I think KnitML can really rock it! Now that there's stitch chart output in KnitML 0.5, most knitters will finally see that KnitML is not just for computer geeks playing with XML :-))
BTW: I think it would be great to have a graphical stitch editor inside GPEC ...
Have you read "Knitting for Anarchists" by Anna Zilboorg? She actually shares your vision: Every knitter should knit according to her own style!
And with KnitML the utopia of knitting anarchy might finally become reality.
Personally I think most of the problems you mention stem from the fact that western hand knitting notation is action-based rather than structure-based.
Imagine flatknitting, roundknitting, top-down. bottom-up, left handed, english, continental, knitting with two needles, several needles, with a knitting loom or different kinds of knitting machines.
Every tool and technique would require a slightly different knitting instruction, and writing transformers to convert between each pair of those formats would quickly become really nasty.
But if you code for fabric structure rather than the process of how to create the fabric, this could serve as a universal intermediate format from which all action-based formats could be derived by transformation ...
Japanese knitting instructions might well serve as an inspiration since they rely heavily on structure based stitch charts. Japanese charts even use a standardized knitting font! (JIS L 0201-1995) while the rest of us still have to live in a Bable of non-standard knitting symbols
The japanese symbols for left-leaning and right-leaning decrease are mirror-symmetric, and whether you perform a K2tog or SKPO action depends merely on your reading direction.
So the same visual chart can be used for flat and circular knitting alike (http://www.tata-tatao.to/knit/graphchart/e-chap2.html)
Wouldn't it be great if there was a little japanese knitter inside the KnitML engine, that could easily translate between all those different knitting instructions, needed to make us western knitters happy? ;)
Keep up the good work & stay krafty!
-- Joseph Knecht
P.S.
There's now an article on KnitML in the k2g2 Craftopedia:
http://www.k2g2.org/wiki:knitml