Skip to content

Commit

Permalink
Sym info
Browse files Browse the repository at this point in the history
* Added a box under the Sym mod
* A clearer link to the Forum post
* A link to Getreuer's Sym layer guide.
  • Loading branch information
DreymaR committed Jan 16, 2025
1 parent 86915c4 commit 101e49f
Show file tree
Hide file tree
Showing 3 changed files with 110 additions and 24 deletions.
71 changes: 68 additions & 3 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -9,13 +9,14 @@ Web pages for [dreymar.colemak.org](https://dreymar.colemak.org)

TODO/WIP for the BigBag Forum & web content (mostly for DreymaR's use):
-----------------------------------------------------------------------
- Training page:
- Mention EnterTrained next to CopyType under Type-Reading
- Add a section on burst training; mention 15 s MT tests, and BurstType. Emphasize varied training.

- Alt layout writeup!
- Make it the first section under Layers. Update other sections to be more layout inclusive.
- Move the links to Getreuer's and the AKL beginner's guides there from the links page, or keep both?

- There's so much strongly worded nonsense on Reddit etc...! People defend their investments and pet ideas.
- I feel like warning against some bad choices: Dvorak (unless implementation is king), Workman/Norman, Minimak.
- Too little and too much. I think Colemak's both good and easy to use, but I wouldn't settle for less.
- There's an abundance of newer layouts, keeping less QWERTY similarity and using more metrics than Colemak.
- The test of time has shown Colemak to be a well-made layout. Lots of work went into its design and testing.
- There are only a few of the many newer layouts about which similar claims may be made confidently.
Expand Down Expand Up @@ -50,6 +51,70 @@ TODO/WIP for the BigBag Forum & web content (mostly for DreymaR's use):
- Not all of us will ever type really fast!
- If something's mostly noticeable at above, say, 120 WPM and like me you'll probably never get there, what then?
- What you know beforehand will influence your experience.
- I had a Discord chat with Supernova, an interested user. I could use this, more or less! (Edited further here.)
Alpha Base, This Is Enterprise. Do You Copy?
--------------------------------------------
Having been an enthusiastic Colemak user for 18 years since 2007, I sometimes get questions like "Why haven't you changed to something newer? There's been a lot of developments, and newer metrics are now taken into account.".
It's a fair question, sure enough. There's always a pot of gold glistening in the distance!
Well, to be honest I never saw a compelling reason to "upgrade" from Colemak, beyond the CAWS mod battery and my other tricks and tools.
The perceived benefits seemed uncertain, and sure enough: After a few months, many users had found perceived or real flaws and chased on to the next "soup du jour".
In the eye of that storm, Colemak is still a solid layout – even nearly 20 years after its inception.
I have a feeling that too many users, especially younger ones, easily get caught up in pretty endless goose chases and rabbit hole spelunking, and I don't want to contribute overmuch to that.
A layout change is hard work, and if the gains from it are at best minor and at worst uncertain – well, then your time is better spent elsewhere unless you think a change would be major fun.
I honestly believe that there's a lot of misinformation out there, from people who don't try stuff out properly and/or guard their investments and/or are otherwise biased and misinformed.
Places like Reddit are especially plagued with strongly worded opinionating, in my impression. You really shouldn't put stock in a newcomer's soap box agenda after trying a layout for, say, one month or three.
In the Alt Keyboard Layout Community, there's pretty much a consensus that some layouts are way oversold by their creators and followers.
Dvorak, Workman, Norman, and, say, Halmak, still attract new users, when there are much better options. Really. For Dvorak, at least it's preinstalled on all OSes; but so is finally Colemak too.
At the same time, this could just as well be an argument for never leaving QWERTY in the first place, couldn't it? And I've really enjoyed doing that!
What constitutes too little and too much may be individual, but I do believe that QWERTY and various minimalistic layouts like Minimak are too little.
To shine a little light on this confusing subject, I can recommend the two guides I linked to on my links page: Getreuer writes very well, and the AKL intro guide is a great introduction to the field.
Delving Too Greedily
--------------------
As you see from the BigBag front page, the base layout is only the beginning. Extend, special dead keys and ergo mods are my main thing. And I'm having such a great time of it!
- My two thumb keys are a good example: The Repeat key does away with a lot of same-finger load, and the CoDeKey
- The Extend and Extend-tap layers are another excellent addition. I truly believe that you should learn such tools well if you haven't yet, as they will benefit you a whole lot more than chasing base layout perfection could ever do.
- Even if you do use Extend or something similar, getting to know its possibilities even better and optimizing your Ext-tap mappings etc is extremely useful. Trust me! I'm still discovering improvements after many happy years of use.
Still, what about all these new layouts, huh? Huh?
[Box: A Quest For The Gral]
Well, finally my curiosity got the better of me, and as a result I'm trying out one of the newer layouts.
Feels really odd after having typed on Colemak for 18 years, I can tell you!
I'm only around half my Colemak speed on Gralmak (my Graphite variant) yet, but I feel I have an idea of what it's about by now.
So far, my main impression is that both Gralmak and Colemak are good, and I won't be getting any substantial gains from the switch – apart from a satisfied curiosity and the fun and pain(!) of trying something new.
Colemak-DH is a good and well thought-out layout. So is Graphite. The latter takes some factors into consideration that weren't yet considered back in 2014. Fine. But I honestly don't feel the differences much (yet?).
They both have their stronger and weaker points. Some musings:
- Graphite has some amazing n-grams, and more than Colemak it uses all three rows for them without much scissoring.
- Colemak has a nice flow to it and some delightful home row rolls, but also some sticky spots here and there.
- Colemak does of course have more redirects than Graphite. Notable examples are YOU, and WAS. More rolls makes for more redirects.
- Writing YOU isn't great on either, just a little easier on Graphite. I've heard layout designers say that it's hard to make that word good on a good layout.
- Writing PHYSICS is worse on Graphite (yes, I'm a phycisist). Every layout has some words that don't agree well with it.
- For some awkward bigrams, alt-fingering helps a lot. Colemak `kn kl` and Graphite `sc sw` are good examples. This takes time to learn well.
- You will get used to a number of weakish spots with time and experience, as long as they aren't too common.
- If you start overthinking such weak spots though, I believe you won't be able to judge a layout fairly. I think this happens to many.
- It won't matter to all, but Colemak is still easier to learn with more QWERTY and shortcut similarities.
- Colemak keeping ZXCV in place is an asset to some, and so is the Tarmak learning method for gradual learning.
As for Graphite vs the very similar Gallium, I tried both and Gralmak ended up with a sort of mix between them and some Colemak principles.
- Like Colemak, I keep the symbol keys unchanged, instead using my Sym mod and special keys. This makes it easier to learn, too.
- I prefer J in the middle and apostrophe on the side, like Gallium and Colemak have. This links in with the first point.
- The only other difference between Gal and Gra is CV vs WZ. I first wanted to go with Gal to make the SC bigram easier, but now I'm on board with Gra keeping W on top and CV on lower row.
- This is mainly because those are more in line with what we're used to. In the long run, that's not a strong argument, but it makes transitions easier.
- The bigrams involved are fine either way, really.
There is no undisputed king of keyboard layouts, and there won't be. You get to a point where individual preference and usage patterns start to matter more than objective aspects, and then you can't agree on a one-layout-fits-all.
I believe we're already about there, pretty much. New tech may yet change typing, but as it stands the main barrier to speed is mental load so a really good development should make it simpler, not more complex.


- Platform box for Sequencing? And Modifiers?
- Colemak-eD too?
Expand Down
Loading

0 comments on commit 101e49f

Please sign in to comment.