A (mildly insane) proposal for new keysets

[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.