geekhack

geekhack Community => Keyboard Keycaps => Topic started by: NoPunIn10Did on Wed, 28 August 2019, 11:36:40

Title: [SURVEY RESULTS] Ergodoxian & Ortho Keycap Label Preferences
Post by: NoPunIn10Did on Wed, 28 August 2019, 11:36:40
Ergodoxian & Ortho Keycap Label Preferences

UPDATE: This survey is now closed.  See full report in replies below.
Raw Survey Results (https://forms.gle/UR3zuFoHqLvWVpKB7)

Do you use either an Ortho keyboard or an Ergodoxian keyboard (Ergodox, Ergodash, and similar)? Would you like to? If so, what labels would you want to have available for your keycap set?

I created the linked survey (https://forms.gle/UR3zuFoHqLvWVpKB7 (https://forms.gle/UR3zuFoHqLvWVpKB7)) to try and gather some data on this topic. My goal is to share the results here and on Geekhack, with the intent of providing better information about what labels Ergo / Ortho users prefer on their caps.

Background

It can be difficult and/or cost-prohibitive to produce keycap kits for Ergodoxian users. Likewise, Ortho users are sometimes lumped into a "40s" kit that only partially represents the labels they'd like on their keyboard. While these tend to be less of an issue for uniform profiles (like DSA or SA R3), it can be frustrating for designing sculpted sets (GMK, SA, DSS, etc.).

The hope here is that the data collected from this survey will be used by keycap designers in their interest checks and group buys.
Title: Re: [SURVEY CLOSED] Ergodoxian & Ortho Keycap Label Preferences
Post by: NoPunIn10Did on Tue, 03 September 2019, 11:40:33

Update 03 Sept 2019

I stopped receiving new replies over the last couple days, so I've closed the survey with 58 responses.

You can follow the original survey link to see the results for yourself, and I'm planning on posting a short summary of the results both here and on reddit.
Title: Re: [SURVEY CLOSED] Ergodoxian & Ortho Keycap Label Preferences
Post by: NoPunIn10Did on Tue, 03 September 2019, 14:53:44
SURVEY RESULTS & COMMENTARY


Before We Begin

Before we get into the details of the responses, a few caveats.  This was not in any way a scientific survey.  Respondents were all volunteers.  I didn't prevent duplicate responses, but I also didn't see any in the results, so I'm assuming that no one responded more than once.

I'd like to note that several respondents stated their preferences either for blank modifiers or for generic-but-distinguishable labels that don't assign specific keys to specific functions.  This is clearly one possible solution for providing Ergo & Ortho-friendly keycaps. 

The goal of this survey was to determine where agreement already exists in regard to sculpted keycap labels.  Designing kits that will actually hit MOQ is clearly a challenge; kits must provide enough coverage for high compatibility without providing so much redundancy as to yield an unpalatable price point.  I sincerely hope keycap set designers will use this information to make determinations regarding which caps to include, which caps to exclude, and which caps are so varied that they might be best served by a blank or novelty legend.

About the Responses

(https://i.imgur.com/F1S2dLXm.png) (https://imgur.com/F1S2dLX)

There were 58 total responses.  Within those, there was a slight bias toward Ortho users over Ergo users.

(https://i.imgur.com/sx7AjMhm.png) (https://imgur.com/sx7AjMh)

~85% of responders either already use or would prefer to use a sculpted keycap set.

QWERTY v. non-QWERTY alphanumerics

(https://i.imgur.com/mLb4Rznm.png) (https://imgur.com/mLb4Rzn)

~71% of responders use QWERTY, QWERTZ, or AZERTY.  ~10% use Dvorak, and about ~12% use one of the Colemak layouts (standard, DH, or DHm). 

I don't know how this compares to the overall population of keyboard users, but I suspect non-QWERTY usage is higher among persons who use Ortho and/or Ergo keyboards.

Takeaway: ~30% Ergo/Ortho users use Non-QWERTY
While non-QWERTY layouts likely need to be a separate kit, they are worth considering for keycap sets supporting Ergo/Ortho keyboards.

Bottom Row Profile

(https://i.imgur.com/VGZHiz7m.png)  (https://imgur.com/VGZHiz7)(https://i.imgur.com/W0NGSXrm.png) (https://imgur.com/W0NGSXr)

When it came to bottom row profiles (R4 or R3), Ergo and Ortho responses fell into approximately the same distribution.  Around a quarter of responders prefer R3 for the bottom row, about a third prefer R4, and the rest prefer R4 for GMK and R3 for SA.

Bottom Row Profile Takeaway: No Clear Agreement
If you're designing kits for the bottom row of Ortho / Ergo keycaps, you might want to include both an R3 and an R4 version.  If that's not going to be cost effective, use R3 for cap profiles similar to SA and R4 for profiles similar to GMK. 

There were additional comments that spacebars should be convex rather than concave, which is another consideration (though such molds may not always be available).

Bottom Row Legends

(https://i.imgur.com/SJDowCkm.png) (https://imgur.com/SJDowCk)

As for which labeled keys to include in the bottom row, there were a few distinct patterns.  In the above chart, I've separated responses that were available selections in the survey versus those that were provided in the "Other" option selection.

Bottom Row Legend Takeaways
In my opinion, based on these responses, I would consider including the following legends:
Because these legends were frequently input in the "Other" section, I'd consider including them as well:

Because bottom row keys are more numerous than side modifiers, there is likely a bit more room for including some redundant keys.  Particularly in the case of the Ergodox, unused bottom-row legends could also come in handy in the thumb clusters (which were not studied directly in this survey).

Regarding "Highly Popular" and "Moderately Popular"
I am not a professional statistician, so I'm using a fairly rudimentary method for identifying mod popularity in the categories below.  For each grid of responses in the sections below (1u Ortho, 1.5u Ergo, et cetera), I calculated the 25th (Q1), 50th (median), and 75th (Q3) percentiles for response count of each legend/row combo.  Using a standard formula for determining upper outliers (1.5*(Q3-Q1) + Q3), I calculated the Upper Fence of the data set.

Legend/row combos whose response counts were equal to or greater than the upper fence are considered "Highly Popular."  Other legend/row combos in the upper quartile (greater than Q3 and less than the upper fence) are flagged as "Moderately Popular."

In the spreadsheet screenshots seen below, Highly Popular combos are shaded green, while Moderately Popular combos are shaded yellow.

1u Modifiers for Ortho
(https://i.imgur.com/OVofoyd.png) (https://imgur.com/OVofoyd)

Highly Popular 1u Ortho Legends

Moderately Popular 1u Ortho Legends

Please note that a few of the moderately popular modifiers are keycaps that would normally be included in a base alphas kit.

Ortho Mod Takeaway 1: Provide Base Alphas Separately
First, I would like to address the issue of alphanumerics.  Some GBs provide a generic base alphanumeric kit, roughly 45 keycaps.  Some provide a base kit including all alphas and modifiers for a standard TKL.  Others provide Ortho and/or Ergo kits, but with 40 of the base alphanumerics already included.

Of the two sets that exclude TKL modifiers, I would highly recommend using the approach wherein there's a set of generic base alphas (and Ortho mods are sold as a separate kit rather than an all-in-one).  R1 Backtick, R1 equals, and R1 minus keys were all found to be moderately popular, and the R2 bracket keys had nearly enough positive responses to be moderately popular as well.  All five of these legends are already included in most standard base alphanumeric kits.

Ortho Mod Takeaway 2: Most Common Mods
Setting aside concerns regarding the bottom row (addressed earlier in this post), side modifiers of an Ortho kit should likely include at minimum all of the Highly popular keys identified above: R2 tab, R3 quote, R4 Alt, and R4 Shift.  Beyond that, GB designers may wish to include some of the moderately popular legends listed above, though including all of them may prove cost-prohibitive.

As noted elsewhere, including blank, novelty, or generic legends may also be a cost-effective option for providing coverage on those keys that are not more highly agreed upon.

1.5u Modifiers for Ergo
(https://i.imgur.com/OfgS9iN.png) (https://imgur.com/OfgS9iN)

Highly Popular 1.5u Ergo Legends

Note that both Backtick and Escape are popular legends for R1, but they typically compete for the same position (left-top mod) rather than being used simultaneously.

Moderately Popular 1.5u Ergo Legends

Ergo 1.5u Mod Legend Takeaways
Many Ergo modifier kits already take the approach of including generic or blank modifiers; this may continue to be a popular means of Ergo coverage.  However, it does seem likely that the Highly Popular legends listed are all safe inclusions in an Ergo kit.  It may be a sound approach to include these highly popular legends instead of only blanks.

One thing that surprised me was the level of agreement on the bracket modifiers; they were the only vertical 1.5u keycaps of any popularity.

A lower number of responses were received in this part of the survey as compared to the 1u Ortho section, so one should take these results with a grain of salt. 

1u Modifiers for Ergo
(https://i.imgur.com/8vbXhj5.png) (https://imgur.com/8vbXhj5)

This section of the survey should have been much better-structured.  I realized after publicizing the survey that outside the R1 modifiers, it might not be entirely clear where the modifiers in the 1u Ergo grid would actually go.  For R3 or R4, are these the bottom-row legends (redundant with another part of the survey)?  Were the respondents assuming the 1u keycaps in this grid might be added to the thumb clusters?

Given that level of ambiguity, I really only want to highlight two R1 legends: = + and - _.

Ergo 1u Mod Legend Takeaway
On a standard Ergo keyboard, there are only two 1u keycaps considered "modifiers" that are outside the bottom row or thumb clusters.  Often these are generic or blank in kits.  As addressed in the Ortho 1u section, providing a full set of alphanumerics as a base kit (rather than a reduced set of 40) automatically grants users two of the most popular 1u modifiers: equals and minus.

Note that this takeaway excludes bottom-row considerations, which are addressed in an earlier section.

Ergo 1u Top Row Colorway Frustration (Anecdotal)
This takeaway was not directly studied in this survey, but it is a point of frustration I'd like to address.
Many Ergo kits seem to assume that the top row 1u keys follow a standard sort of arrangement (with M standing for modifier):
1 2 3 4 5 M M 6 7 8 9 0

This assumption is not necessarily correct.  Anecdotally, I can confirm that Ergo users may adopt a variety of different top-row layouts. 
Examples:
1 2 3 4 5 6 7 8 9 0 M M
M 1 2 3 4 5 6 7 8 9 0 M
1 2 3 4 5 6 M 7 8 9 0 M


This ends up being a point of potential annoyance when an Ergo kit's colorway adopts different colors for its top-row 1u mods than it assumes for its number keys.  Such an assumption can lead to undesirable aesthetic outcomes for users not adopting the assumed "default" layout.  To make it easier on said users, consider providing the 1u top-row modifiers in the same colorway as the number keys, thereby granting much more aesthetic flexibility to the order of that row's legends.

2u Keys for Ergo
(https://i.imgur.com/0PTv9YF.png) (https://imgur.com/0PTv9YF)

This is another section of the survey that I realized was a bit flawed after posting it.  The difference between a left or right thumb-cluster Ergo keycap is mostly meaningless, so it was not sensible to separate those categories.  So, for measuring the median and quartile responses on this grid, I first combined Left and Right into a single group: Either.

In the responses, only the Blank / Space keycap was highly popular.  The following vertical legends were moderately popular as part of either a left or right thumb cluster:

I also found moderate agreement that a Ctrl key would not be used as one of the vertical 2u keys.

(https://i.imgur.com/cCdKIml.png) (https://imgur.com/cCdKIml)

However, I found the results from the question regarding the preferred shape of keycaps to be much more interesting.  For this question, I asked, "Which shapes of 2u vertical keys would you use in your thumb clusters?" and provided the following options:
As seen in the chart above, the first and fourth options received the highest number of responses.  "Flat" vertical 2u keys, similar to SA R3, are clearly a popular option.  However, an equal number of responses stated that they'd like Ergo kits to include multiple 2u profiles, even if they are blank.  Such an approach would also satisfy responders who marked options #2 or #3 above.

Ergo 2u Vertical Keycap Takeaway
The safest approach for vertical 2u keys for an Ergo kit is to leave them all blank.  Instead of providing redundancy and variation in the legends, however, it is advisable to provide multiple shape options for Ergo users to try.  Blank slanted keycaps can satisfy users who wish to have either an upward-sloping or downward-sloping vertical key.

Overall Conclusions & Additional Notes
In the final open-ended section of the survey, I had several participants note they preferred generic or blank modifiers.  The data collected here shows that high agreement exists for some legends, however.  An approach worth consideration might be a mixed one: to provide most or all of the highly-popular legends in an Ergo/Ortho kit alongside blank or generic keycaps to fill out the remaining modifier positions.  Such a mixed approach could yield some redundancy without the highly inflated keycap count required to provide all moderately popular legends.

Other participants mentioned, both within the survey and in comments around the survey announcements, that having Ortho modifiers provided solely with 40s kits can be very frustrating.  The survey did not address this frustration, but I expect it arises partly because such kits often only include a narrow subset of popular Ortho mods and don't always include generic or blank modifiers to satisfy more custom setups.  Per the data seen in the Ortho 1u Mod section, only a handful of such mods are highly popular, so including only a small set of specific legends (without any generic options) is likely to lead to a less-than-satisfactory result.

I also had several participants note that they were Boardwalk users.  For those unfamiliar, a Boardwalk keyboard is one using primarily Ergodoxian keycaps in an arrangement that happens to fit inside many popular 60% keyboard cases.  The one point where Boardwalk users may differ from Ergo users in keycap needs is in the bottom row.  Some Boardwalk layouts use standard spacebars and modifiers from TKL sets in their bottom row.  Others use the 2u vertical keys as split spacebars, but placed horizontally instead. 

It is not entirely clear that a standard Ergo kit could ever fulfill all Boardwalk users' needs, but I would definitely recommend including at least two "flat" 2u keys (like SA R3) as options for vertical thumb-clusters (so that they might be used horizontally instead).
Title: Re: [SURVEY RESULTS] Ergodoxian & Ortho Keycap Label Preferences
Post by: NoPunIn10Did on Mon, 09 September 2019, 12:27:42
I have finished writing up the above report and would be interested in any further feedback that the Geekhack community might be able to provide.
Title: Re: [SURVEY RESULTS] Ergodoxian & Ortho Keycap Label Preferences
Post by: deacon on Mon, 09 September 2019, 16:10:52
Thanks for writing this up, maybe it will help. At least I know I'm not the only one who prefers blanks.

FWIW, the Boardwalk fits in 60% cases.
Title: Re: [SURVEY RESULTS] Ergodoxian & Ortho Keycap Label Preferences
Post by: harlekein on Mon, 09 September 2019, 19:12:29
Excellent survey and write up. It's not surprising to me that ortho users more regularly use an alternative layout to QWERTY as well, since they're already willing to break away from the norm.

Furthermore, I do really think providing GMK base plus 40s/ortho set is really a disservice instead of a service. Ortho users aren't really being catered to with that and we shouldn't accept it if we want this changed.
Title: Re: [SURVEY RESULTS] Ergodoxian & Ortho Keycap Label Preferences
Post by: NoPunIn10Did on Mon, 09 September 2019, 19:55:36
Thanks for writing this up, maybe it will help. At least I know I'm not the only one who prefers blanks.

FWIW, the Boardwalk fits in 60% cases.
Ah, good catch. I’ve corrected that note in the writeup.
Title: Re: [SURVEY RESULTS] Ergodoxian & Ortho Keycap Label Preferences
Post by: NoPunIn10Did on Mon, 09 September 2019, 20:01:26
Excellent survey and write up. It's not surprising to me that ortho users more regularly use an alternative layout to QWERTY as well, since they're already willing to break away from the norm.

Furthermore, I do really think providing GMK base plus 40s/ortho set is really a disservice instead of a service. Ortho users aren't really being catered to with that and we shouldn't accept it if we want this changed.

I do understand the desire to group layouts together to make MOQ easier to reach, but I really think the better grouping would be Ortho + Ergo (Ortho/Dox, if you will). There seems to be much more overlap: the bottom row is quite similar, flat keycaps in the thumb clusters could work in an Ortho context as well. It’s really only the 1.5u mod keys that are a waste for Ortho users.
Title: Re: [SURVEY RESULTS] Ergodoxian & Ortho Keycap Label Preferences
Post by: harlekein on Tue, 10 September 2019, 04:30:16



I do understand the desire to group layouts together to make MOQ easier to reach, but I really think the better grouping would be Ortho + Ergo (Ortho/Dox, if you will). There seems to be much more overlap: the bottom row is quite similar, flat keycaps in the thumb clusters could work in an Ortho context as well. It’s really only the 1.5u mod keys that are a waste for Ortho users.

I agree. 40s just has too many oddities which are no use to ortho. Look at KAT milkshake.

Combining ergo and ortho at least results in extended ortho coverage for boardwalk or other thumb clusters.