A (mildly insane) proposal for new keysets

[VinnyCordeiro]

Organizing a keyset group buy isn't the easiest thing of the world, far from that. Each country have its own standard layout, even between countries sharing the same language (France & French Canada, Portugal & Brazil, Spain & Spanish Latin America, etc).

That affects directly the availability. In a niche market as this, making only the most common layouts is a matter of keeping costs as low as possible. That have a side effect: the uncommon layouts are not supported OR are even more expensive to be acquired.

I've made an empirical study, comparing some of the most common national keyboard layouts (not restraining myself to mechanical keyboards) : United States (ANSI), United Kingdom (ISO), Germany (ISO QWERTZ), France (ISO AZERTY), Italy (ISO), Spain (ISO), Norway (ISO), Sweden/Finland* (ISO) and Brazil (modified ISO, called ABNT2).

^{*The layouts of these two countries are similar enough to be considered the same. Feel free to correct me.}

The result is here: http://www.keyboard-layout-editor.com/#/gists/a3b0575e0b8432a31a28

This link shows only the 60% portion of the keyboards, the one that have most differences (numpad differences are negligible and not treated here). The colored keys are those that changes between layouts, the white ones are common to all of them.

All that said, my proposal is this: make all group buys from now on with a mandatory base set (the white keys on the study) and a mandatory national set (the colored keys on the study). That way we reduce the price paid for all users that are buying ISO sets (being honest here: US is the market for almost anything in the world, it is understandable that ANSI layouts are the standard keysets for this reason), because they would not subsidize keys that they won't use. (Example: Granite International Keycap Set, $35.99 option at Massdrop containing freaking 86 keycaps when people would use just a handful of them.)

The intention here is to reduce overall costs, but I could be wrong. Opinions are welcome, but let's keep this discussion sane and civilized.

[SpAmRaY]

Depending on keycap manufacturer all legends/correct profile may not be available.

Besides that you can't force people to have specific caps in a group buy if they don't want to, that is just silly.

[jdcarpe]

I have a better proposal: everyone in every country standardize on the US ANSI keyboard layout. Simple, and cost effective.

[lolpes]

Why not have the alternate letters for multiple languages doubleshoted/dyesubed/pad printed on different locations of the caps with the same color? Kinda like the dolch set has support for DE language by side printing the alternate versions on the Y and Z caps.

[VinnyCordeiro]

Depending on keycap manufacturer all legends/correct profile may not be available.

Agreed. That would be a problem with doubleshot keycaps, but dyesubbed ones are not restricted to that.

Besides that you can't force people to have specific caps in a group buy if they don't want to

And why would they? No one is proposing that. My idea is: you buy 'base set + US' to have a US ANSI keyset. If you live in France, you buy 'base set + FR' and have a nice AZERTY keyset. All these options are still dependent on MOQ limits imposed by manufacturer.

[FrostyToast]

Are people really that bothered by having legends that aren't in their language?
People here should know how to touch type or take the time to learn.

[VinnyCordeiro]

I have a better proposal: everyone in every country standardize on the US ANSI keyboard layout. Simple, and cost effective.

Unfortunately not gonna happen.

[lolpes]

I have a better proposal: everyone in every country standardize on the US ANSI keyboard layout. Simple, and cost effective.

If we all spoke the same language and used the same characters for writing then yes, but that is not the case. And there is a simple cost effective way, use blanks.

[VinnyCordeiro]

Are people really that bothered by having legends that aren't in their language?
People here should know how to touch type or take the time to learn.

On all keyset group buys threads there are always people complaining about this. I'm thinking about those people. And touch type, while desirable, is a skill that very few people develops, so legends are still necessary.

[VinnyCordeiro]

If we all spoke the same language and used the same characters for writing then yes, but that is not the case. And there is a simple cost effective way, use blanks.

Blanks are cost effective, I agree. But only if you touch type and that isn't true for a great part of keyboard users.

[SpAmRaY]

Depending on keycap manufacturer all legends/correct profile may not be available.

Agreed. That would be a problem with doubleshot keycaps, but dyesubbed ones are not restricted to that.

Besides that you can't force people to have specific caps in a group buy if they don't want to

And why would they? No one is proposing that. My idea is: you buy 'base set + US' to have a US ANSI keyset. If you live in France, you buy 'base set + FR' and have a nice AZERTY keyset. All these options are still dependent on MOQ limits imposed by manufacturer.

Ok that makes more sense however you won't ever reach MOQ on anything but base + US

But seriously nobody would buy the base set if the 'country' set they want isn't going to be made.

[azhdar]

I have a better proposal: everyone in every country standardize on the US ANSI keyboard layout. Simple, and cost effective.

why should the us be picked above other layouts?


@Op I'll always be supportive of international keysets offering.
The thing is this kind of thing is highly dependent on the manufacturer. For example GMK, an international kit would end up costing about 80-100$ for a MoQ of 250, and I think very few people are ready to drop this ammount of money to have 1/10 or less of the kit used.

It's a possibitity with Signature plastic with their lower MoQ, we have seen it happen in granite and in the recent SA Retro.

I though about it for a while and for me the best option to make it happens is to dyesub keycaps. Because unless I'm wrong you could print like 250 ANSI, 50  ISO QWERTZ Kits, 20 ISO AZERTY Kits ... without a problem and each individual will pay for what he desires.

[azhdar]

Are people really that bothered by having legends that aren't in their language?
People here should know how to touch type or take the time to learn.

I know a lot of ISO user makes the transition into the UK ISO keysets or even towards ANSI, but some of us can't.
Weirdly enough I like my keyboard to type what is printed onto the keys.

[jbondeson]

Sadly while there is a vocal minority of people who wish to see international layouts included in buys, the reality is that when it comes time for people to put up with their wallets it always falls short, even for really small MOQs. The only reason why Granite worked was because DSA is row agnostic so you don't have as many permutations as a contoured set.

So unless you do a mega-set with international support baked into the base kit (raising the price considerably) ISO users are basically screwed. 

[My_Thoughts]

I have a better proposal: everyone in every country standardize on the US ANSI keyboard layout. Simple, and cost effective.

We will get right on that as soon as you start using A4 paper, logical date formats (YYYY-MM-DD or DD-MM-YYYY we can work with either) and the metric system 

[baldgye]

The only way it can work, is the way Bunny has done it with the Bro-set and Miami Twice. You buy a set and that set includes a full ANSI and ISO set.
I can understand the pain of people from non-English language counties, my Italian SO suffers like mad with my boards because they have the UK ISO layout, but at least that layout is possible, if I had been forced into ANSI it would have been painful for both of us.

As is I think having US ANSI and UK ISO is the best compromise as it works and fits two distinct layouts, but I wish this people were more open minded to helping support others, I still remember the amount of **** Ivan had to wade through when he tried to include other languages etc...

[lolpes]

If we all spoke the same language and used the same characters for writing then yes, but that is not the case. And there is a simple cost effective way, use blanks.

Blanks are cost effective, I agree. But only if you touch type and that isn't true for a great part of keyboard users.

The last part was meant as a joke i understand the topic at hand and i am not a touch typist sadly

[FrostyToast]

The US layout will always reach moq before any others because it is the most standard layout.
Even the Asian countries largely use US ANSI.
Now when I say that there are several countries that use one layout, and that some key sets can barely reach that moq, do you really think that a layout only popular in a single country will reach moq?

[My_Thoughts]

I've made an empirical study, comparing some of the most common national keyboard layouts (not restraining myself to mechanical keyboards) : United States (ANSI), United Kingdom (ISO), Germany (ISO QWERTZ), France (ISO AZERTY), Italy (ISO), Spain (ISO), Norway (ISO), Sweden/Finland* (ISO) and Brazil (modified ISO, called ABNT2).

Thanks for doing this work, it could be very useful in the future.   The UK layout is interesting for keycap designers.  Most seem to add the 3 symbols to the first key in the first row (left of the 1) but designers have different views on the 4 key.  Apart from Apple I have not seen a commercial UK keyboard without $ and € on the 4 in a loooong time.  Some keycap designers swap the $ for the € and leave the $ off all together.  That said I think most people here type € more than $.   For designs that cannot get 3 symbols on the keys I really like the 4€ and not the 4$

[jdcarpe]

I have a better proposal: everyone in every country standardize on the US ANSI keyboard layout. Simple, and cost effective.

We will get right on that as soon as you start using A4 paper, logical date formats (YYYY-MM-DD or DD-MM-YYYY we can work with either) and the metric system 

Fine by me. And while we're at it, let's get rid of time zones and daylight saving time. Such a dumb concept, really. Everyone on the planet can just use UTC, please. Oh, and that reminds me...why are we still calling these objects the Earth, Sun, and Moon. Can't we decide to call them Terra, Sol, and Luna, instead? Earth is the ground, and there are lots of moons in the solar (hey!) system, and many suns in the galaxy.

[baldgye]

Fine by me. And while we're at it, let's get rid of time zones and daylight saving time. Such a dumb concept, really. Everyone on the planet can just use UTC, please.

omg yes please

[tofgerl]

I have a better proposal: everyone in every country standardize on the US ANSI keyboard layout. Simple, and cost effective.

Make Obama suggest it and the EU will accept it within thirty minutes

[jonathanyu]

I have a better proposal: everyone in every country standardize on the US ANSI keyboard layout. Simple, and cost effective.

doesn't really work while typing Japanese. typing Japanese on jis layout is much easier imo.

[jdcarpe]

I have a better proposal: everyone in every country standardize on the US ANSI keyboard layout. Simple, and cost effective.

doesn't really work while typing Japanese. typing Japanese on jis layout is much easier imo.

I'm sure that's true. And it follows that typing other languages might be easier on their ISO language keyboards. But it's going to take quite some time to get everyone to use American English exclusively. Maybe standardizing on the US ANSI keyboard will be the catalyst for change!

[jbondeson]

We will get right on that as soon as you start using A4 paper, logical date formats (YYYY-MM-DD or DD-MM-YYYY we can work with either) and the metric system 

I think this is a fair trade as long as you go YYYY-MM-DD, because going least significant first is just odd.

[tofgerl]

Look, it's quite easy to find out if this will work: Someone organizing a groupbuy should make one full base+ansi set available, in addition to all the sets Vinny suggest, excluding the ansi set, since no one would order the base and ansi set separately instead of the combined set.
So the list would look like this:
1. Ansi TKL/60%/Whatever GB organizer wants
2. International base
3. UK pack
4. DE pack
5...

Now, if not enough people buy the international base, and you really want the ISO pack and the base, you can hope for your international pack to tip as well as secure the normal ANSI pack.

If only the ANSI pack tips, then we know it won't work. But sitting in this thread and saying it won't work is just pessimism rooted in ignorance.

Edit: It's a little like a 7bit buy, but with only 2% of the options.

[SpAmRaY]

Look, it's quite easy to find out if this will work: Someone organizing a groupbuy should make one full base+ansi set available, in addition to all the sets Vinny suggest, excluding the ansi set, since no one would order the base and ansi set separately instead of the combined set.
So the list would look like this:
1. Ansi TKL/60%/Whatever GB organizer wants
2. International base
3. UK pack
4. DE pack
5...

Now, if not enough people buy the international base, and you really want the ISO pack and the base, you can hope for your international pack to tip as well as secure the normal ANSI pack.

If only the ANSI pack tips, then we know it won't work. But sitting in this thread and saying it won't work is just pessimism rooted in ignorance.

Edit: It's a little like a 7bit buy, but with only 2% of the options.

I'm thinking Ivan tried something like this and it didn't work out.

[jbondeson]

Multiple people have tried.

One of the biggest issues with a child kits is that people who need the kit are loathe to commit to the buy until the child kit they want looks certain to tip, so you end up with this chicken and the egg issue where people need to commit to tip, but don't want to commit prior to tipping...

This is where software could help out a little bit so that the GB organizer wouldn't have to figure out all out.

[HoffmanMyster]

I have a better proposal: everyone in every country standardize on the US ANSI keyboard layout. Simple, and cost effective.

We will get right on that as soon as you start using A4 paper, logical date formats (YYYY-MM-DD or DD-MM-YYYY we can work with either) and the metric system 

Fine by me. And while we're at it, let's get rid of time zones and daylight saving time. Such a dumb concept, really. Everyone on the planet can just use UTC, please. Oh, and that reminds me...why are we still calling these objects the Earth, Sun, and Moon. Can't we decide to call them Terra, Sol, and Luna, instead? Earth is the ground, and there are lots of moons in the solar (hey!) system, and many suns in the galaxy.

Sounds good to me too. I'm already using the proper ISO-approved date format, so that one's done. I'm perfectly familiar with the metric system and obviously prefer it. I'm fully prepared to ignore all daylight savings changes, and I can easily add 6 hours to my clock. 

How are things progressing over there, My_Thoughts? 

[zslane]

It costs nothing but the organizer's time to offer all those international child kits. After all, the ones that don't reach MOQ simply don't get made. At least they were offered, which is all any GB organizer can hope to do.

However, it only takes one GB where the organizer goes to all the trouble to put together seven ISO child kits that don't tip to convince him not to bother ever again. And it only takes one savvy organizer who learns from the experiences of others to not bother in the first place.

[alexjd99]

I would think it would be better for someone else to organize a group buy, and distribute only what is needed, so that people who live in place X, only get keycaps for place X, and someone in place Y only gets keycaps from place Y

Not sure if that's the best way to explain it, but I'm sure someone will know what I mean

[suicidal_orange]

Looking at these layouts it's very hard to work out what the designers were thinking - does anyone use < often enough to make it a non-mod key?  Technically it is surely my native UK-ISO along with the few BR users that are at fault for not having the "standard" <> key, but what a crazy standard.  Same with Germans being the only layout with just Z and Y switched - why be awkward?  You could rearrange the letters within their rows if you use one letter less (I don't remember their being many Vs in German, why not put Z there?)  At least the French went completely mad and moved lots of letters but surely A shouldn't be way up in the corner, and needing a mod for a full stop?!  It's no wonder azhdar et al. lament the lack of AZERTY sets

The only key I'd contest is the E € one - would anyone really cry if the € was missing?  It's such a sensible place for the symbol you can't really lose it...

If we're going for a crazy GB why not make each key that changes a separate item so the many QWERTY users can share the Y with the ANSI people, while the French and Germans can share their Z and SWI/FI and DE users share Ö and Ä etc making these keys cheaper.  Once the main buy is closed and everyone knows what's tipped there could be another round for blanks to fill in the gaps - you may not know exactly where the key you're after is, but it's down there somewhere and it's obviously not those caps as they have legends, so it must be the blank.  Or if MoQ was low enough buying 5 of a couple of keys to get a complete set would hardly break the bank...

Or what about running this with a handful of buyers who are willing to buy multiple sets of their "local" layout safe in the knowledge that there will be future enthusiasts who will buy them?  Would probably require an American to hold all the money as importing lots of sets from SP would surely attract insane customs charges.  Though I couldn't really wish the nightmare that the sorting would be on anyone that would be the best way.  If only I lived in the US...

[Oobly]

The only way a GB with multiple international packs works is if it's massively popular (which you won't know beforehand, so... problem).

The best way to manage this whole thing, IMHO, is to have a GB system that allows fallback kits to be ordered. So you order a Finnish / Swedish kit with a generic ISO (like a UK kit, but possibly with some extra legends) as a fallback. The system has an option for "If this kit doesn't tip, add this kit to my order instead:"
That way people won't be scared to order, since they know they won't be stuck with a set that doesn't even physically fit their board.

The way Vinnie proposed won't work because if your particular kit doesn't hit MOQ you're stuck with an incomplete set that you can't even put on an ANSI board or sell to anyone.

It costs nothing but the organizer's time to offer all those international child kits. After all, the ones that don't reach MOQ simply don't get made. At least they were offered, which is all any GB organizer can hope to do.

However, it only takes one GB where the organizer goes to all the trouble to put together seven ISO child kits that don't tip to convince him not to bother ever again. And it only takes one savvy organizer who learns from the experiences of others to not bother in the first place.

It's not just time. Depending on the kit, some new legends may need cutting. With the current GB systems I'd most likely only offer a small "generic" ISO kit. Big ISO kits cost the buyer too much when they're not even going to use most of the caps in it. Individual country ones don't hit MOQ for various reasons.

[zslane]

Oh, definitely.

It is one of the drawbacks of being a member of a very small minority. You can't benefit from quantity production. What you want will always cost a lot more than the standard sets that get ordered in the hundreds/thousands. It is simply a fact of life.

I think if someone doesn't want to pay the premium that goes with their niche layout, they have a choice: opt out of most keycap sets, or switch to a standard layout that guarantees a more affordable buy-in. You can't have it both ways.

Now, in theory you could have a second-tier GB where a central entity buys a whole bunch of "All The ISOs In One" kits and then parcels out the subsets according to who needs what. Unfortunately, the overlap between ISO standards for certain keys means too many buyers will be in competition for the same common keycaps, so even that strategy doesn't really work.

[suicidal_orange]

Was my suggestion just too insane to consider or were people put off by the length of the post and my insulting of our forefathers?

@Oobly we're going dyesub due to the many new legends - this idea has enough problems without waiting/paying for molds!

[My_Thoughts]

I think the major issue for Euro ISO users is there seem to be very few people from each country into this sort of thing.  There are a few Italians/Germans/Scandinavians here, on reddit and deskthority but not enough to tip every GB.  Then depending on what is purchased the postage cost is another barrier.  I tend to think if the community expanded to include many more Europeans then child deals would drop all the time.

Many people I know speak 2 languages and often 3.  Everyone seems to have their own way of dealing with different characters and I suspect that impacts on what deals sell as well.
 

[My_Thoughts]

I have a better proposal: everyone in every country standardize on the US ANSI keyboard layout. Simple, and cost effective.

We will get right on that as soon as you start using A4 paper, logical date formats (YYYY-MM-DD or DD-MM-YYYY we can work with either) and the metric system 

Fine by me. And while we're at it, let's get rid of time zones and daylight saving time. Such a dumb concept, really. Everyone on the planet can just use UTC, please. Oh, and that reminds me...why are we still calling these objects the Earth, Sun, and Moon. Can't we decide to call them Terra, Sol, and Luna, instead? Earth is the ground, and there are lots of moons in the solar (hey!) system, and many suns in the galaxy.

Sounds good to me too. I'm already using the proper ISO-approved date format, so that one's done. I'm perfectly familiar with the metric system and obviously prefer it. I'm fully prepared to ignore all daylight savings changes, and I can easily add 6 hours to my clock. 

How are things progressing over there, My_Thoughts? 

I showed my German girlfriend an ANSI keyboard last night and she spent some time talking about the inefficiency of the smaller enter key.  Then I thought about the money I have spent on ISO keyboards and I don't think it's going to happen - sorry.  but at least we both can use metric

[My_Thoughts]

Another thought is maybe one reason the ISO sets don't drop is for people not using the UK layout they need to buy base + "ISO" (usually enter + UK keys) + their local layout + extras.

I wonder if putting the ISO enter into the main set, and then have several Euro sets without the enter would work.  UK + De perhaps.  Scandinavian set etc.  Or package the enter with each of the child sets.  Yes the price would go up a little, but the non UK people would not be buying the UK keys they don't need.

[jdcarpe]

Would an ISO Enter, a few blanks in the appropriate row profiles, and sub-legends to accommodate QWERTZ and AZERTY work for most people? That could probably be included with every ANSI set at little extra cost for everyone.

[SpAmRaY]

Would an ISO Enter, a few blanks in the appropriate row profiles, and sub-legends to accommodate QWERTZ and AZERTY work for most people? That could probably be included with every ANSI set at little extra cost for everyone.

But don't people usually say they don't want blanks with a printed set?

[azhdar]

Would an ISO Enter, a few blanks in the appropriate row profiles, and sub-legends to accommodate QWERTZ and AZERTY work for most people? That could probably be included with every ANSI set at little extra cost for everyone.

That wouldn't work for me

[jdcarpe]

Would an ISO Enter, a few blanks in the appropriate row profiles, and sub-legends to accommodate QWERTZ and AZERTY work for most people? That could probably be included with every ANSI set at little extra cost for everyone.

That wouldn't work for me

Please elaborate.

[azhdar]

Would an ISO Enter, a few blanks in the appropriate row profiles, and sub-legends to accommodate QWERTZ and AZERTY work for most people? That could probably be included with every ANSI set at little extra cost for everyone.

That wouldn't work for me

Please elaborate.

Like Ray said blanks in the middle of printed keys looks awful imo.
And I'm not sure of what u meant by

sub-legends to accommodate QWERTZ and AZERTY work for most people?

But I'd do with a full set of blank anyday rather than "incorrectly" (aka not the layout used) printed

[My_Thoughts]

Would an ISO Enter, a few blanks in the appropriate row profiles, and sub-legends to accommodate QWERTZ and AZERTY work for most people? That could probably be included with every ANSI set at little extra cost for everyone.

That wouldn't work for me

When buying custom keycaps we are already paying a lot for what we get (they are small demand custom products)  When I spend over $100 just on a set of caps I really want them all printed and not a few blank ones

[jdcarpe]

Like Ray said blanks in the middle of printed keys looks awful imo.
And I'm not sure of what u meant by

sub-legends to accommodate QWERTZ and AZERTY work for most people?

But I'd do with a full set of blank anyday rather than "incorrectly" (aka not the layout used) printed

Like this:

Would an ISO Enter, a few blanks in the appropriate row profiles, and sub-legends to accommodate QWERTZ and AZERTY work for most people? That could probably be included with every ANSI set at little extra cost for everyone.

That wouldn't work for me

When buying custom keycaps we are already paying a lot for what we get (they are small demand custom products)  When I spend over $100 just on a set of caps I really want them all printed and not a few blank ones

Oh I see. Well, US ANSI users could subsidize a few "special" keys, but not with legends for every language. Now we see why language packs never make MOQ.

[azhdar]

Like Ray said blanks in the middle of printed keys looks awful imo.
And I'm not sure of what u meant by

sub-legends to accommodate QWERTZ and AZERTY work for most people?

But I'd do with a full set of blank anyday rather than "incorrectly" (aka not the layout used) printed

Like this:

Show Image

That would work for me.

What we want is quite simple: have the key types what's printed on it :o

[jdcarpe]

What we want is quite simple: have the key types what's printed on it

I understand that. I'm sure you also understand that isn't possible for 100% of people, with every language layout, and a limited amount of keys which can be included in a group buy to make MOQ.

[azhdar]

What we want is quite simple: have the key types what's printed on it

I understand that. I'm sure you also understand that isn't possible for 100% of people, with every language layout, and a limited amount of keys which can be included in a group buy to make MOQ.

I'm 100% aware it's hard if not impossible to have international keysets massproduced, you'll notice I never push the issue in IC thread or GroupBuys.

But if we want to do it it's pointless to do half of the job with blanks. Imo it's either 100% correct or not done at all.
There have been International kits done in SP sets where the third legend (AltGr layer) wasn't printed, it's as incorrect than blanks to me.

The way matt3o has done Granite and SA Retro is the correct way.

[ideus]

I have a better proposal: everyone in every country standardize on the US ANSI keyboard layout. Simple, and cost effective.

If we all spoke the same language and used the same characters for writing then yes, but that is not the case. And there is a simple cost effective way, use blanks.

We all know that there is a way to write most languages using US International, that is using an US ANSI keyboard to type with Alt Gr characters most Latin based languages, the exceptions being those languages like Russian, Chinese, Korean and all that requires a full set of different characters.

[Melvang]

What about subsidizing the cost for language packs into the main set?  A couple dollars from each set should more than pay for some language packs.

Remember, SP technically doesn't have an MOQ.  They will make you a custom one off set.  Granted it will cost you a significant chunk of change but they will do it.  I think I saw numbers between $700 and $800 for a one off set.

I don't think most people would have an issue of throwing a couple dollars to help support language packs.

[jbondeson]

What about subsidizing the cost for language packs into the main set?  A couple dollars from each set should more than pay for some language packs.

Remember, SP technically doesn't have an MOQ.  They will make you a custom one off set.  Granted it will cost you a significant chunk of change but they will do it.  I think I saw numbers between $700 and $800 for a one off set.

I don't think most people would have an issue of throwing a couple dollars to help support language packs.

I think that makes sense on DSA where bouncing a key around works, but if you look at the SA Retro packs you had 37 keys in the DE-NO-SE kit, and 43 in the FR-IT kit. 

No matter what we'd need to find some kind of compromise that the ISO people would be willing to live with on at least a few of those keys.

[jdcarpe]

What about subsidizing the cost for language packs into the main set?  A couple dollars from each set should more than pay for some language packs.

Remember, SP technically doesn't have an MOQ.  They will make you a custom one off set.  Granted it will cost you a significant chunk of change but they will do it.  I think I saw numbers between $700 and $800 for a one off set.

I don't think most people would have an issue of throwing a couple dollars to help support language packs.

I think that makes sense on DSA where bouncing a key around works, but if you look at the SA Retro packs you had 37 keys in the DE-NO-SE kit, and 43 in the FR-IT kit. 

No matter what we'd need to find some kind of compromise that the ISO people would be willing to live with on at least a few of those keys.

Exactly. If we are trying to cut costs to make it more reasonable, the US ANSI users could subsidize some subset of the international language packs, but not even close to all of them. Some people are going to have to compromise, but no one wants to. Every buyer wants HIS language pack to be the one included with the main set. 80 additional keys is not feasible. Get it down to 10 or so, and we can work the numbers.

Doing a GB like the OP suggests, with a small main pack, and kits for every language, is a sure way to fail. Maybe 7bit could pull it off, but doubtful.

[Melvang]

What about subsidizing the cost for language packs into the main set?  A couple dollars from each set should more than pay for some language packs.

Remember, SP technically doesn't have an MOQ.  They will make you a custom one off set.  Granted it will cost you a significant chunk of change but they will do it.  I think I saw numbers between $700 and $800 for a one off set.

I don't think most people would have an issue of throwing a couple dollars to help support language packs.

I think that makes sense on DSA where bouncing a key around works, but if you look at the SA Retro packs you had 37 keys in the DE-NO-SE kit, and 43 in the FR-IT kit. 

No matter what we'd need to find some kind of compromise that the ISO people would be willing to live with on at least a few of those keys.

Exactly. If we are trying to cut costs to make it more reasonable, the US ANSI users could subsidize some subset of the international language packs, but not even close to all of them. Some people are going to have to compromise, but no one wants to. Every buyer wants HIS language pack to be the one included with the main set. 80 additional keys is not feasible. Get it down to 10 or so, and we can work the numbers.

Doing a GB like the OP suggests, with a small main pack, and kits for every language, is a sure way to fail. Maybe 7bit could pull it off, but doubtful.

Pretty much this.  There is no way to effectively reduce the cost on individual language packs outside of manufacturers lowering their costs.  The only way to do that is to just have a larger buy and more packs are sold.  But if the cost for small productions numbers is offset into the cost of the base kit, that can drastically reduce the cost for the pack for the few people that need them.  Think of it this way.

If a cost without any subsidy is $50 for a 4 key kit, and only 10 kits are needed, but you are selling 500 base sets, a $1 premium into each base kit would totally fund that kit.  I feel that this is more than reasonable for the buyers of any base kits.  Yes they are paying for keys they will not be receiving, but say the base kit is $60, that would only be a cost increase of 1.6% for the base kit.  Totally reasonable, and something I (and probably others) would be more than willing to help fund.

[azhdar]

I typed something about dyesubbing above.
I'm not sure how dyesubber usually work but couldn't a buy happen where 600 base sets are produced and then each indivual language kit is produced in lower numbers?

[Melvang]

I typed something about dyesubbing above.
I'm not sure how dyesubber usually work but couldn't a buy happen where 600 base sets are produced and then each indivual language kit is produced in lower numbers?

Legend type doesn't make much of a difference unless new legends need to be cut for double shot which is ~$25 from SP iirc.

[zslane]

New SA double-shot plates now cost $50 to have cut. And that's per keycap mold that it applies to. So if you need a new legend to be available in row 2 for one language, but row 4 for some other language, then you get to pay that $50 twice. Funding lots of new double-shot legends for fully sculpted SA keycaps is simply not economically feasible in most cases. 7bit manages it because his GBs are so huge he can spread the cost of new legends over enough kits that nobody notices the cost increase for subsidizing them. That would be a lot harder to pull off when you only have a few kits in your set to absorb those costs.

I'm not sure how I feel about seeing international users standing with their hats in hand asking for their special keycaps to be subsidized by the buyers of the dominant standard. For a lot of US buyers it isn't about how much or little it would add to their cost, but the principle of increasing the price of 95% of the base sets to accomodate the needs of 5% of the customer base. I'm just not sure how popular such a move would be.

[Niomosy]

I think the more popular sets would be able to offset those costs easier.  You'd need a better system in place to handle that kind of thing though.  MassDrop certainly couldn't handle it; they've got problems galore with how they handle child kits already and don't even have anything resembling Oobly's fallback idea in place.

For smaller runs, you probably can do it.  Would US ANSI users be fine with it?  I can't speak for everyone but a few bucks extra for me is fine if it helps others out but everyone has their limit on that cost-wise.

[FrostyToast]

I think the more popular sets would be able to offset those costs easier.  You'd need a better system in place to handle that kind of thing though.  MassDrop certainly couldn't handle it; they've got problems galore with how they handle child kits already and don't even have anything resembling Oobly's fallback idea in place.

For smaller runs, you probably can do it.  Would US ANSI users be fine with it?  I can't speak for everyone but a few bucks extra for me is fine if it helps others out but everyone has their limit on that cost-wise.

A few bucks extra is what we already pay to cover for the UK ISO sets.
Would I pay a few dollars extra on top of that to support JIS or whatever language is next? No.
People should learn to accept that they can't have most sets in their language and at least be happy they have the specific layout they need.

[livingspeedbump]

Who would enforce these "mandatory" base kits? PMK sure won't, Massdrop sure won't. I generally am a fan of the kits the CTRLalt guys throw together, seems to cover a lot of stuff, without really going overboard, but don't see them doing this.

I mean, quiet frankly $35 bucks isn't bad if you really feel the need to have all those labels perfectly correct on your keyboard. Personally, I'll keep doing the international kits like I always have, weighing the pricing and the kit sizes careful to make something affordable, yet covering the largest percentage of people I can at the same time. I'm not going to make all the ANSI buyers pay for for ISO keys though when they don't need them though.

[Niomosy]

I think the more popular sets would be able to offset those costs easier.  You'd need a better system in place to handle that kind of thing though.  MassDrop certainly couldn't handle it; they've got problems galore with how they handle child kits already and don't even have anything resembling Oobly's fallback idea in place.

For smaller runs, you probably can do it.  Would US ANSI users be fine with it?  I can't speak for everyone but a few bucks extra for me is fine if it helps others out but everyone has their limit on that cost-wise.

A few bucks extra is what we already pay to cover for the UK ISO sets.
Would I pay a few dollars extra on top of that to support JIS or whatever language is next? No.
People should learn to accept that they can't have most sets in their language and at least be happy they have the specific layout they need.

Fair enough.  I do agree that it does become questionable to keep bumping the base price to accommodate a small number of users, though I try to be somewhat flexible about it  Given the nature of international languages and keyboard layouts, you can't please everyone. 

[Oobly]

New SA double-shot plates now cost $50 to have cut. And that's per keycap mold that it applies to. So if you need a new legend to be available in row 2 for one language, but row 4 for some other language, then you get to pay that $50 twice. Funding lots of new double-shot legends for fully sculpted SA keycaps is simply not economically feasible in most cases. 7bit manages it because his GBs are so huge he can spread the cost of new legends over enough kits that nobody notices the cost increase for subsidizing them. That would be a lot harder to pull off when you only have a few kits in your set to absorb those costs.

I'm not sure how I feel about seeing international users standing with their hats in hand asking for their special keycaps to be subsidized by the buyers of the dominant standard. For a lot of US buyers it isn't about how much or little it would add to their cost, but the principle of increasing the price of 95% of the base sets to accomodate the needs of 5% of the customer base. I'm just not sure how popular such a move would be.

Legends can be used across row profiles, but not cap sizes, so a Row 2 legend can be used on a Row 4 cap, but only if the caps are the same sizes.

IMHO, it depends on the popularity of the sets, but if Massdrop (or PMK) could implement the "fallback" system I think that's the best for all. You could actually then even remove the duplicated caps from the base kit and have an ANSI kit, too. At least that one will be guaranteed to hit high order numbers. Nobody has to pay for caps they don't use, everyone gets the right caps for their language if their kit tips and if it doesn't they at least still get caps that will fit the physical layout of their board.

It's true that SP doesn't have a set MOQ, but for most kits the cost at levels below 25 is very high. You have to quote some kind of figure on the GB for each kit and people will simply not order if the quoted price is too high,, locking it to be a high cost, low orders kit, catch-22.

Even better could be something like Massdrop's "commit at this level" AND the fallback option for each kit. Then even kits that don't hit 25 will be made for those brave souls who've commited at low numbers, but they'll cost them a lot. If it doesn't hit your commit level, you get the cheaper fallback kit.


@suicidal_orange: I like the concept behind your idea, but it has three flaws:
1. logistics nightmare - I would not like to be the one to sort all the individual cap kits to orders. Bound to go wrong and and have at least a few incorrect caps / missing caps sent out.
2. PMK limit of 10 kits, but this only affects the GB if you run it through PMK.
3. Assuming one or two of your caps doesn't hit MOQ... A separate blanks GB afterwards is a hassle to arrange and will need a new production slot (which may even create a subtle change in the colour). And I for one don't want an almost complete Finnish kit with blanks. I'd rather have a fully complete UK kit.

TLDR: GB systems need to be upgraded to include "commit at this level" with "fallback" option per kit.

[tofgerl]

Frankly it makes sense to me, being an international person with a very international brain that international kits should be more expensive, since they're made in fewer numbers.
That's why I use ANSI. That, and I'm a developer, and developing in ****ing Javascript when I need to do shift+altgr+8 to get a simple { almost causes me to go into fetal position...

[tofgerl]

Frankly it makes sense to me, being an international person with a very international brain that international kits should be more expensive, since they're made in fewer numbers.
That's why I use ANSI. That, and I'm a developer, and developing in ****ing Javascript when I need to do shift+altgr+8 to get a simple { almost causes me to go into fetal position...

But it's not all bad for ISO Norwegian - we have a special HTML/XML key with both < and > on it! Yay, too bad I use Jade and JSON...

[Oobly]

Frankly it makes sense to me, being an international person with a very international brain that international kits should be more expensive, since they're made in fewer numbers.
That's why I use ANSI. That, and I'm a developer, and developing in ****ing Javascript when I need to do shift+altgr+8 to get a simple { almost causes me to go into fetal position...

Yup. I was frankly shocked to discover the Finnish / Swedish layout as a programmer, since there are so many good coders from these countries, but their board layouts truly SUCK for coding.

[livingspeedbump]

Frankly it makes sense to me, being an international person with a very international brain that international kits should be more expensive, since they're made in fewer numbers.
That's why I use ANSI. That, and I'm a developer, and developing in ****ing Javascript when I need to do shift+altgr+8 to get a simple { almost causes me to go into fetal position...

Yup. I was frankly shocked to discover the Finnish / Swedish layout as a programmer, since there are so many good coders from these countries, but their board layouts truly SUCK for coding.

I just wanted to throw out that I was born in Espoo 

[Niomosy]

Frankly it makes sense to me, being an international person with a very international brain that international kits should be more expensive, since they're made in fewer numbers.
That's why I use ANSI. That, and I'm a developer, and developing in ****ing Javascript when I need to do shift+altgr+8 to get a simple { almost causes me to go into fetal position...

I think I would go crazy if I had to do that for curly brackets.  I go through a decent enough amount in Puppet code and even with the Eclipse add-ons handling the closing brace for you when you create the opening one, I'd still go nuts.

[Oobly]

Frankly it makes sense to me, being an international person with a very international brain that international kits should be more expensive, since they're made in fewer numbers.
That's why I use ANSI. That, and I'm a developer, and developing in ****ing Javascript when I need to do shift+altgr+8 to get a simple { almost causes me to go into fetal position...

Yup. I was frankly shocked to discover the Finnish / Swedish layout as a programmer, since there are so many good coders from these countries, but their board layouts truly SUCK for coding.

I just wanted to throw out that I was born in Espoo 

I was born in Cape Town, South Africa, but have been living in Tampere, Finland around 9 years now. Both are amazing places and I feel blessed to have been able to live in both. I didn't realise until I got here that I was really a Finn at heart and it felt like coming home.

[DanielT]

As a non-english native speaker I can share my 0.02$ on international layouts.
For my language (and it's not German) nobody bothered to make a special keyboard except for a few very rare rubberdomes and I think I saw a couple of times some crappy Cherry MY boards, this goes even deeper, nobody bothered to build a decent localization character set, some of the letters are wrong in the current implementations, they look like the real thing but are wrong, and even many of my countrymen complained and sent correct specifications nobody bothered. What was the result ? Well, we adapted, we are very good at this, except for books/newspapers and some really official documents nobody and I mean nobody uses localization. We all know how the read the words correctly and there is no confusion.
All my boards are ANSI US layout, I use the US-International for German but even there localization can be avoided.
As a long time UNIX user (not LINUX, I mean the old-school UNIX ****) for me there is only one locale:

Code: [Select]

$ locale
LANG=C
P.S: I want to be noted that I'm not an ANSI-US extremist, and I don't have anything against other layouts I just don't see a point in spending the extra money for something I have never used and will never use. I fully understand why others don't feel like me, I just though that an opinion from a non-english speaker would be good to have.

[n__dles]

Well, we adapted, we are very good at this, except for books/newspapers and some really official documents nobody and I mean nobody uses localization. We all know how the read the words correctly and there is no confusion.

The problem arises when not everyone has adapted. For the languages I speak, and I'd imagine most Western languages, there's no intelligibility problems when only using the English alphabet. However, at a minimum missing the additional letters/modifiers makes you stand out, and in formal settings is unacceptable.

I found this thread while searching for "US International layout" as an alternative to my current method. As I write I prepend missing characters/accents with punctuation. Then I run a script to replace them with the appropriate character, so "a,c~ao" becomes "ação".

For my language

Which I'm very curious to know, if you don't mind sharing. Guesses are: a language which is not spoken by the majority of any country, something like Romansch; or one where the alphabet has been changed, like in Transnistria.

We will get right on that as soon as you start using A4 paper

zomg A4 is my shiz!! So expensive and hard to get in the US. I'm going to start a We the People petition for a switch to A4 paper. Where's my change Obama!!

[jamster]

Would an ISO Enter, a few blanks in the appropriate row profiles, and sub-legends to accommodate QWERTZ and AZERTY work for most people? That could probably be included with every ANSI set at little extra cost for everyone.

That would work for the one board that I have that's some weird-ass 1800 AZERTY thing. When I bought this particular board, I realised it would be hard to find caps for, and would be just as happy using blanks for the alphas. It's only the symbols on the numeric row that I have trouble touch typing reliably.

[jamster]

Are people really that bothered by having legends that aren't in their language?
People here should know how to touch type or take the time to learn.

On all keyset group buys threads there are always people complaining about this. I'm thinking about those people. And touch type, while desirable, is a skill that very few people develops, so legends are still necessary.

I have thought that touch typing is such a fundamental skill this century that the vast majority of mechanical keyboard enthusiasts who care enough about changing caps would have learnt to touch type. All through my last degree and at any workplace I've been at since, I don't recall anyone who could not touch type. I know a lot of people who don't touch type 'by the book', but they don't have to stare at their keyboards, they just use non-standard fingers to hit keys.

[DanielT]

Well, we adapted, we are very good at this, except for books/newspapers and some really official documents nobody and I mean nobody uses localization. We all know how the read the words correctly and there is no confusion.

The problem arises when not everyone has adapted. For the languages I speak, and I'd imagine most Western languages, there's no intelligibility problems when only using the English alphabet. However, at a minimum missing the additional letters/modifiers makes you stand out, and in formal settings is unacceptable.

I found this thread while searching for "US International layout" as an alternative to my current method. As I write I prepend missing characters/accents with punctuation. Then I run a script to replace them with the appropriate character, so "a,c~ao" becomes "ação".

I use this keyboard layout, it's the US International with AltGr not the other one with dead keys, it's good enough for almost all the languages, except mine...

For my language

Which I'm very curious to know, if you don't mind sharing. Guesses are: a language which is not spoken by the majority of any country, something like Romansch; or one where the alphabet has been changed, like in Transnistria.

There's no secret, I'm Romanian, we used to have Cyrillic alphabet until sometime in the 18-19th Century, back then the schools were ran by the Orthodox church and they used this alphabet, when we started to open more towards the western civilisation we also adopted the Latin alphabet. yet, we still have a couple of extra characters that are unique to our language Ț,ș,â,ă ,î . Unfortunately the US International doesn't cover these.
Because in the 90's after the fall of Communist Regime all computers that were imported had ANSI layout , or almost all them, and no OS had support for our language we adapted to that. I remember back in the 90's I had a friend who wrote the code for Romanian language support, I used that when I wrote my UNI diploma It took a long time to have support for our language in mainstream OS's, but it the mean time we have all adapted to English/US so that only very few people (only legal, official and news papers/books) use the Romanian character set. I have to admit that this is a shame, you loose part of your identity ...
I would love to have a keyboard with Romanian characters printed on the keys, but except for some crappy rubberdomes and some really rare Cherry lasered boards there's no hope for that. We are a really small minority in the keyboard world that no one will make this a reality. 

[keyboardia1]

Are people really that bothered by having legends that aren't in their language?
People here should know how to touch type or take the time to learn.

I don't mind using the wrong legends for my language on my hhkb. but when i shell out a lot of money for a nice looking keyset, i want it to show the right legends. i mean, that's kind of the point when you're buying a keyset with legends on it ...

[DanielT]

Are people really that bothered by having legends that aren't in their language?
People here should know how to touch type or take the time to learn.

I constantly see this false assumption that everyone can touch type or should learn to do so. The reality is that not everyone can. I can touch type, it's not a problem, but if I'm tired or in a hurry my brain stops to coordinate my hands, it's a small problem I have and can't be helped, in that case legends help a lot.
And as others have said, if we "all can touch type" why the hell do we run GB's with nice colourways and cool legends ?

[n__dles]

There's no secret, I'm Romanian, we used to have Cyrillic alphabet until sometime in the 18-19th Century, back then the schools were ran by the Orthodox church and they used this alphabet, when we started to open more towards the western civilisation we also adopted the Latin alphabet. yet, we still have a couple of extra characters that are unique to our language Ț,ș,â,ă ,î . Unfortunately the US International doesn't cover these.
Because in the 90's after the fall of Communist Regime all computers that were imported had ANSI layout , or almost all them, and no OS had support for our language we adapted to that. I remember back in the 90's I had a friend who wrote the code for Romanian language support, I used that when I wrote my UNI diploma It took a long time to have support for our language in mainstream OS's, but it the mean time we have all adapted to English/US so that only very few people (only legal, official and news papers/books) use the Romanian character set. I have to admit that this is a shame, you loose part of your identity ...
I would love to have a keyboard with Romanian characters printed on the keys, but except for some crappy rubberdomes and some really rare Cherry lasered boards there's no hope for that. We are a really small minority in the keyboard world that no one will make this a reality.

Whoa  , I'm kinda shocked that I didn't know that there's not a Romanian layout and that Cyrillic was used at some point. Here's why:

• I'm interested in languages, especially Romance languages.
• I'm like when languages switch to different alphabet families, like in Tranistria and especially Judaeo-Spanish.
• I spent several years in Madrid, where Romanians are the largest foreign resident group.
• and especially because a dear friend of mine, who I've known for some 15 years is a Romanian linguist

Cyrillic being used a couple centuries ago because of the Orthodox church isn't such a surprise. IIRC in Transnistria it's not related to the church, but politically motivated; establish sovereignty / curry favor with Russia.

I suppose that even missing those letters, Romanian is still littered with hats and tails, so to the uninitiated it doesn't look like anything is missing. I also have the impression that Romanians, especially those with money, buy things like electronics abroad for better selection/prices.

I use this keyboard layout, it's the US International with AltGr not the other one with dead keys, it's good enough for almost all the languages, except mine...

Show Image

It would be nice to find quality key caps in US International, but I think what I'll probably do is print the layout and leave it on my desk for reference. ATM I only need to use the extras for formal Spanish stuff, which I write a handful of times a month.

I handwrite notes a lot more. For the fun of it, I want to learn the Hebrew alphabet for to write Spanish in it. Judaeo-Spanish was written in it, so I can just follow their orthography. I've had a look and keycaps in Hebrew are virtually non-existent, I guess stickers are my best option. Anyhow this is very long and far off the original topic.. Multumesc for the information  :thumb:

[jdcarpe]

Cyrillic being used a couple centuries ago because of the Orthodox church isn't such a surprise. IIRC in Transnistria it's not related to the church, but politically motivated; establish sovereignty / curry favor with Russia.

I suppose that even missing those letters, Romanian is still littered with hats and tails, so to the uninitiated it doesn't look like anything is missing. I also have the impression that Romanians, especially those with money, buy things like electronics abroad for better selection/prices.

I don't get the need for diacritics. Russian Cyrillic can be transliterated to straight Latin characters, so why do Slavic languages which use the Latin alphabet feel the need to include them?

I do think that English should have its digraphs removed. If we need a new symbol for those sounds, let's do that. Not a fan of diacritics, obviously.

* jdcarpe asks that you please excuse the nerdy linguist talk here.

[VinnyCordeiro]

Cyrillic being used a couple centuries ago because of the Orthodox church isn't such a surprise. IIRC in Transnistria it's not related to the church, but politically motivated; establish sovereignty / curry favor with Russia.

I suppose that even missing those letters, Romanian is still littered with hats and tails, so to the uninitiated it doesn't look like anything is missing. I also have the impression that Romanians, especially those with money, buy things like electronics abroad for better selection/prices.

I don't get the need for diacritics. Russian Cyrillic can be transliterated to straight Latin characters, so why do Slavic languages which use the Latin alphabet feel the need to include them?

I do think that English should have its digraphs removed. If we need a new symbol for those sounds, let's do that. Not a fan of diacritics, obviously.

* jdcarpe asks that you please excuse the nerdy linguist talk here.

You don't get the use of diacritics because English rarely use them, usually on loan words; the native English words that do have diacritic signals are rarely written with them anymore.

Fortunately English isn't the only language spoken/written in the world. It isn't even the most spoken language in the world if that matter, but I digress.

Just because you don't see the point in something it doesn't mean it isn't important. Different languages have different uses of different written signals, and those signals are important in that context.

[tofgerl]

Well, this thread took a left turn somewhere

[VinnyCordeiro]

Well, this thread took a left turn somewhere

I agree, and because of that I'm locking it. Unfortunately.