In the school I went to, 2 years was the expectation with graduation requirements being 240 WPM with 98% accuracy.
How quickly you actually got there was very much dependent on the person. Some do it quicker than 2 years with little practice, some do it slower than 2 years with a lot of practice, and some appeared to be unable to reach the requirements at all.
I don't really remember too well at this point but it maybe took me a few months to get to 30 - 40 WPM, maybe a year to get to 120, and year and a half in total for 240.
I don't think it's practical or worth it for the vast majority of people to learn stenography to be honest. That being said, Plover is a nice and much needed initiative, I'm glad it exists.
It was at a small accredited school which offered associate degrees in court reporting (also degrees in a handful of other subjects such as accounting). I had to complete a few other courses not strictly related to stenography, but the degree itself was an associates degree in court reporting.
That's a very healthy qwerty pace. It would many months, perhaps years, of daily practice before you exceed that. I've been lurking in that community and unless you have A LOT of prose to produce which can justify the up-front learning investment, steno is IMHO best approached as neat hobby/game that also happens to get you fast text input ... eventually. Else you are in for a frustrating wait.
Oh, and to your question, very good stenographers head off into the stratosphere at 200wpm+ (elite in 300wpm+) for extended periods, rates not really doable on qwerty except for short burst of predictable prose IMO.
From what I have read, you can go to over 150 in 6 months. Of course, it would depend on how much time you spend on it.
My biggest concern is that there seems to be no standard for programming. I have seen some people develop their own cords, but that seems like quite a time investment to me.
Another think that might be an issue is if you have to be able to type in different languages. I guess you could adapt easily enough as long as the language already has a chord-table, but it also means more time learning.
> My biggest concern is that there seems to be no standard for programming.
The standard would definitely have to vary by language, and you would probably define custom strokes for each defined symbol in the source, similar to the strokes for English words in ordinary steno. It could be workable if the stroke definitions could be stored with the original source code as custom comments, and then imported by your stenotype tool.
Mapping dot and Enter to thumb keys on an ergonomic keyboard does wonders in combination with this. I actually have both mapped to the same key (in different layers).
If your IDE is good at predicting which class member you will type, it can be lightning fast.
People often say things like "I think slower than I type", but I don't think that this is correctly modeling the problem, because it's only true over long timescales.
Consider this: the average American uses only 7 GB of mobile data per month. Would they be happy with 50 kilobits per second? That's much faster than their average usage rate, so in some sense, the data rate wouldn't be the limiting factor.
It's very much the same with typing speed. Yes, your internal word generation rate might be slower than you type if we're looking at an hour by hour, or even minute by minute scale. But once you have a thought crafted, it's beneficial to get it out as quickly as possible.
