Historical question - when did rubber domes eat the world?

micmil

19 May 2021, 21:22

Oh yes, "only 40." So anyone younger than you can't possibly know a fucking thing, right?

Sorry that your anecdotal bullshit doesn't align with actual facts. I point out, AGAIN, that none of our statements are contradictory and that the only thing you're incorrect about is the timeline of when shit really hit the fan.

Then again if you're the ESR I'm thinking you are, it's pretty clear you've never been big on reality unless it directly involves you.

User avatar
Muirium
µ

19 May 2021, 21:34

Micmil, c’mon give it a rest with the ad hominem will ya? I’m about the same age as you and can see he has a point. The keyboard technology turnaround this (needlessly aggro) thread is asking about was well under way when me and you were learning to ride our trainer bicycles and thought Michael Jordan, Micheal Jackson and Michael J Fox were the coolest dudes on earth. Certainly the coolest Michaels. :lol:

If you had a time travelling DeLorean toy when you were a kid (raises hand!) then this happened before your watch. Unless you really were into mechs when your kindergarten friends were into Transformers!

User avatar
raoulduke-esq

19 May 2021, 21:41

Why is every thread turning into a flamewar by the time it gets to page 2 these days?

User avatar
fohat
Elder Messenger

19 May 2021, 23:43

micmil wrote:
19 May 2021, 21:22

doesn't align with actual facts.
What are these "actual facts"?

Aren't we discussing a phenomenon that took the better part of a decade to unfold, with tails on both ends?

Are we looking for the year that rubber dome sales were "x" percent of keyboard sales and all other species were "100-x" percent?

Personally, I would be curious to know what "x" equates to "eating the world"?

Findecanor

20 May 2021, 00:24

esr wrote:
17 May 2021, 19:50
And Compaq passed IBM in sales in '94.
I think the comparison to look for would be IBM vs. everyone else doing PCs, and in number of units more than in dollars. That would likely set the date back a few years.
micmil wrote:
17 May 2021, 21:50
Rubber domes have the added benefit of being super quiet.
I think it can sometimes be hard to distinguish between when rubber domes were purposely engineered/chosen because they were quiet and when "They are super-quiet" was used merely as a sales-pitch.
... And not all rubber-dome keyboards were actually particularly quiet.
esr wrote:
18 May 2021, 11:27
And there were other period keyboards with an artificial click generated through a speaker because the switches were too quiet.
Researchers in human-computer/machine interaction in the 1970's (or earlier) had recognised the value of typing feedback, and this knowledge was widespread way into the '90s.
I think audial feedback was the primary reason for speakers in keyboards, and far from all such keyboards were as loud as the Model M.

I believe that their influence and belief in tactile feedback could also have been the reason why rubber dome switches displaced switches with linear feel in the early '90s, which had before been just as common in cheaper keyboards.

Another question is why more expensive terminals and workstations also started using rubber dome keyboards in the '90s.
DEC chose mushy rubber domes for the VT320 already in 1987. In 1991, Sun introduced the Sun Type 5 which was mushy. In the mid-90's, both SGI and DEC used NMB rubber domes that had "OK" feel. HP also went from Fujitsu leaf-spring to Fujitsu domes.

I wonder if the belief in tactile feel could have affected standards. We know that DIN standards had previously made computers grey/beige and lowered the height of keyboards ...

User avatar
zrrion

20 May 2021, 01:00

I've read through the ISO standards that came after DIN, and feedback of some kind is indeed required. you can either have a beeper or clicky switch for audible feedback or have tactile feedback. If you have a beeper there has to be some way to turn it off. Domes being favored over somehting like spring over membrane is likely due in part to the fact that domes add required feedback with the fewest and cheapest parts.

User avatar
Bjerrk

20 May 2021, 06:12

raoulduke-esq wrote:
19 May 2021, 21:41
Why is every thread turning into a flamewar by the time it gets to page 2 these days?
It seems to come down to only a few (inexplicably confrontation seeking) individuals, such as micmil.

Findecanor and zzrion, you seem to be on to something, that changing standards may at least have played a role in the pre-eat-the-world proliferation of domes!

micmil

21 May 2021, 02:06

Muirium wrote:
19 May 2021, 21:34
Micmil, c’mon give it a rest with the ad hominem will ya? I’m about the same age as you and can see he has a point. The keyboard technology turnaround this (needlessly aggro) thread is asking about was well under way when me and you were learning to ride our trainer bicycles and thought Michael Jordan, Micheal Jackson and Michael J Fox were the coolest dudes on earth. Certainly the coolest Michaels. :lol:

If you had a time travelling DeLorean toy when you were a kid (raises hand!) then this happened before your watch. Unless you really were into mechs when your kindergarten friends were into Transformers!
:roll:

Because the entirety of historical research is only undertaken by people old enough to have been there, right?

Fuck off with that braindead boomer-bitch mentality.

Better idea. I'll do the fucking off and leave you twats to your circlejerk.

User avatar
an_achronism

21 May 2021, 02:49

This place isn't short on entertainment, is it?

User avatar
fohat
Elder Messenger

21 May 2021, 03:06

an_achronism wrote:
21 May 2021, 02:49
This place isn't short on entertainment, is it?
I don't think that anybody has ever topped this gem from Geekhack many years ago:


Quote from: Anonymous Coward circa 2011:

I feel ashamed when I see a grown up man say things like "…. a sharp drop at the actuation point at around 2/3 - 3/4 way down the ...."

Don't you have anything better to do with your life than yammering away on the minute details of a keyboard? I have both an M and an F that I picked up at goodwill for nothing 15 years ago and for the first time yesterday I googled about them and found 'enthusiast' (here an euphemism for retarded) websites where idiots bounce off the walls telling each other about the orgasms per second they have when using them.
And 'using' is an overstatement with 90% of those morons. Most are busy opening them, cleaning the last atom of dirt off them, 'restoring' what doesn't need any restoration, 'upgrading', thinking of names for them, 'modding', taking photos, showing them off, in general jerking off about the clicky sensations and the superb accuracy of their typing and other general uber-dorkiness. What I never found there was anything useful to do with them, ie. actually program a computer.

Go type 'messenger lectures' in youtube and see what smart people look like, then kill yourself disassembling your One True Keyboard (TM) for the nth time and swallowing all the buckling springs.

And then mail one of your remaining model Fs to me.

Quote

User avatar
an_achronism

21 May 2021, 03:22

fohat wrote:
21 May 2021, 03:06
program a computer
Nice

User avatar
raoulduke-esq

21 May 2021, 03:36

Must not have been enough people on Facebook for them to disagree vehemently with that day, so they joined a community dedicated to something they give no shits about to complain about the amount of shits given in said community.

Ain't the interwebz grand?

keyboard Kultist

21 May 2021, 03:59

Hey! I use my vintage keyboards to write programs!!!!

In Fortran.

Because I like it like that.

User avatar
kbdfr
The Tiproman

21 May 2021, 09:05

an_achronism wrote:
21 May 2021, 03:22
fohat wrote:
21 May 2021, 03:06
program a computer
Nice
Quoting ths way lets it appear as if this expression you (nicely) call "nice" had been written by fohat,
when in fact he was himself quoting. Your quote should read:
fohat wrote:
21 May 2021, 03:06
[…] Quote from: Anonymous Coward circa 2011:
[…] program a computer.
[…]
Please note that this remark is in no way intended to be aggressive :mrgreen:

JBert

21 May 2021, 09:14

micmil wrote:
18 May 2021, 04:02
And like it or not, the Black Widow is what you have to "thank" for the modern resurgence of mechanical keyboards but that's a different post for a different topic.
Now that might turn into a heated discussion. The old keyboards weren't quite forgotten in 2010, and Das Keyboard, Filco, Cherry were also not unknown. Better bring some sales numbers when you start that topic. ;)

JBert

21 May 2021, 09:57

So far we had some discussion of how offices might have switched to rubber domes for quieter operation and lower price. However, for the home user it might also have been a step up even when keyboards still were a line item.

I'm too young to really know about the range of personal computers and their keyboards in the 90s, but I remember using my father's Cherry MY in that time and it wasn't all that great. The two springs in there meant that you'd get a somewhat weird force curve, and the keys would bind depending on how you pressed them. I later used a Chicony rubber dome keyboard which was lighter and smoother in its key travel.

Thinking about it, there might be other differences with technology of that age. Compared to buckling rubber sleeves a rubber dome keyboard might have a softer landing because you actually land on the squished dome, rather than crash landing on the contact switch once the sleeve buckles.

The domes can snap, so you get some form of tactility if you just pick the right rubber material (though I should mention that the tactility happens at the wrong point in the key travel). Compare that to foam and foil keyboards where you only get a linear, mushy feeling, or with Cherry MX where they had to use precision-engineered parts to make the slider rub the contacts in just the right way (but at great cost), not to mention the extra number of parts involved in a clicky switch.

So all in all I think that rubber dome keyboards could have made sense even before system vendors started offering them as an included item in the box. It would have been just the middle of the road between operating noise, ergonomics and most of all price.

EDIT: I haven't even talked about chiclet (membrane) keys, though I think that no PC vendor after the IBM PCJr was crazy enough to offer such a cheap keyboard. Anything is an improvement over those.

Findecanor

21 May 2021, 14:20

JBert wrote:
21 May 2021, 09:14
micmil wrote:
18 May 2021, 04:02
And like it or not, the Black Widow is what you have to "thank" for the modern resurgence of mechanical keyboards but that's a different post for a different topic.
...The old keyboards weren't quite forgotten in 2010, and Das Keyboard, Filco, Cherry were also not unknown. ...
I would say that you are both right. Don't forget iOne, SteelSeries and Deck/TG3! I think Deck/TG3 could have been the first to make backlit Cherry keys — for their special-purpose police-car keyboards before they made a full-size desktop keyboard under the Deck brand.

But when Razer introduced the BlackWidow, Razer was already established as a major brand for PC gaming peripherals so they were able to get their keyboards into more stores than the other brands. That made them significant, and what associated the new wave of mechanical keyboards with PC gaming.

BTW. Months before Razer announced it, they had posted an on-line poll asking people which Cherry switch they should use. The mechanical keyboard community of the time — in which fans of Model M and clicky Alps were in majority (or at least the loudest) — voted for the clicky Cherry MX Blue. :P

User avatar
inmbolmie

21 May 2021, 15:21

hellothere wrote:
17 May 2021, 23:41
micmil wrote:
17 May 2021, 21:50
Rubber domes existed before, and had a large market share, but it was that late-90's pricing war that really destroyed all market share that mechanical keyboards had prior.
I think that's about right. In a very general sense, if I see a keyboard with a Windows (95) key on it, I assume it's a rubber dome/membrane unless I have information to the contrary. There were still a bunch of Alps and other switches out there, though. Easy example is the Dell AT101 bigfoot. Black Alps.

Personally, I am interested in when the first rubber dome/membrane non-mechanical came out and what it was. I'm thinking it was probably in the mid 1980s.
The war was lost well before that, but I agree that the bloody Windows 95 key was the cataclysmic event that made instantly obsolete all the previous keyboards in existence, at least for the PC mass market. After that everything was rubber dome or rubber dome cousin.

User avatar
fohat
Elder Messenger

21 May 2021, 16:20

inmbolmie wrote:
21 May 2021, 15:21

the bloody Windows 95 key was the cataclysmic event that made instantly obsolete all the previous keyboards in existence
I had not thought about it in that way, you have an excellent point.

During my decade in the hobby I have conditioned myself to automatically look askance at anything with a Windows key.

esr

22 May 2021, 23:53

fohat wrote:
21 May 2021, 16:20
During my decade in the hobby I have conditioned myself to automatically look askance at anything with a Windows key.
There is some justice in this. On the other hand, I'm typing on a Unicomp New Model M with a Windows key, and it doesn't suck.

I like the heavier case on my Model M 1391401 (1988) better, but I'd I'm not sure I could honestly say the keyfeel on the New Model M isn't just as good. Older Unicomps had iffy build quality but the new tooling has pretty well fixed that.

I'm a little sensitive about this issue, because my preferred window manager is i3 which actually needs the Windows key for production - it's the modifier that tags window manager commands. If I use my 1391401 on my main machine I'm going to have to dive into Soarer's configuration tool and do some remapping.

esr

23 May 2021, 00:58

JBert wrote:
21 May 2021, 09:57
So all in all I think that rubber dome keyboards could have made sense even before system vendors started offering them as an included item in the box. It would have been just the middle of the road between operating noise, ergonomics and most of all price.
Every once in a while I have to remind myself of a maxim I picked up from primate ethology many years ago: "All interesting behavior is overdetermined." "Overdetermined" unpacks here as "has more than one sufficient explanation".

The underlying idea is that in complex adaptive systems evolving under selective pressure - like the gene pool of a species, or the market for computer keyboards 1980-2000 - agents will tend to develop behaviors that solve more than one problem at once, because that's more efficient than having an array of behaviors that solve one problem each.

So just because cost pressure is a sufficient explanation for the rubber-dome disaster, that does not exclude JBert or anyone else being right about other sufficient explanations in the thinking of keyboard vendors. We think these explanations are in competition with each other, but they aren't necessarily - that's just a language bias we have where we tend to assume single effects have single causes.

Sorry, this was a bit meta. My point is that when we're talking about causes of the transition, saying "factor X was enough by itself to drive buckling springs out of the mass market" does not exclude other claims of the form "factor Y was enough by itself to drive buckling springs out of the mass market" for X!=Y. Several claims like that could simultaneously be true.

My judgment, like JBert's, is that cost pressure was the most important factor in the history we observe. But it might be there were other causes that would have killed them off anyway. Caution about jumping to conclusions is wise.
Last edited by esr on 23 May 2021, 04:21, edited 1 time in total.

User avatar
fohat
Elder Messenger

23 May 2021, 03:28

esr wrote:
22 May 2021, 23:53

actually needs the Windows key
Since my preferred daily driver is an ANSI-modified Model F 122-key terminal board, the Windows key (a real one, not a Control-Escape) is no problem. I use the lower row of upper function keys as conventional F1-F12, and the left function keys F1-F8 normally, but turn left F9 into F11 (because I use that often) and left F10 as Windows, only an inch and a half from where it would be otherwise.

esr

23 May 2021, 04:26

fohat wrote:
23 May 2021, 03:28
and left F10 as Windows, only an inch and a half from where it would be otherwise.
I was figuring I'd remap the Pause/Break key.

esr

23 May 2021, 04:32

raoulduke-esq wrote:
21 May 2021, 03:36
Must not have been enough people on Facebook for them to disagree vehemently with that day, so they joined a community dedicated to something they give no shits about to complain about the amount of shits given in said community.

Ain't the interwebz grand?
I dunno. That last line, "And then mail one of your remaining Model Fs to me."? Yeah, that sounded awfully enthusiast, especially for someone who claimed to have an F already.

I feel like I might be reading parody**2 here - Anonymous Coward imitating a downshouter but subverting it in the last line.

esr

23 May 2021, 05:04

fohat wrote:
19 May 2021, 23:43
Personally, I would be curious to know what "x" equates to "eating the world"?
I meant to answer this earlier.

Assuming I could identify it, I would equate x to the year the trend became irreversible.

Discussion on this thread had already left met pretty sure that year x had to be bracketed by IBM's purchase agreement with Lexmark (1991) and Lexmark bailing out of the keyboard business (1996). I previously proposed that it could be pinned to the year Compaq passed IBM in PC sales (1994).

Someone suggested that the year we really wanted would be when aggregate sales of the clonemakers (not just Compaq) exceeded IBM's, and that might be sooner than '94. I think that is a basically sound argument, but one that doesn't move the needle much.

For one thing, not all clonemakers went to rubber domes; I think of the Northgate OmniKey, and I recently learned from one of chyrosran22's reviews that Dell was selling keyboards with ALPS switches at the time. For another thing, Compaq was so huge compared to the other clonemakers that I find it difficult to construct any plausible scenario in which "Compaq dominates" lags "clonemakers dominate" by more than a quarter or two.

Accordingly, for purposes of the video I am comfortable saying "around 1994". This is not far from the 1995 introduction of the Windows key, so it's compatible with hellothere's scenario.

Hak Foo

23 May 2021, 10:49

For my two cents, I feel like by the time Compaq passed IBM, that was largely a statistical novelty. By the 1990s, it wasn't "IBM vs Compaq", it was "every man for themselves". By 1996 (the first figure I could find), neither IBM nor Compaq held even 10% of the market, so they hardly had steering power anymore. Arguably at that point, Intel was the most dominant force-- things like PCI and USB came alongside their new chipset designs, sold by many motherboard makers, so they hit the industry as a wave, more so than "appearing first in (name brand) PC and copied by everyone else".

I feel like IBM lost control of the industry around 1987. First, the Deskpro 386 hit, so they were now followers rather than leaders for the bleeding edge platform-defining aspects. Then, the PS/2 platform landed with a bit of a thud: nice machines, sold well enough, but it forced the rest of the industry to say "we're not going to copy everything IBM does, may as well follow the people who put a new CPU in without demanding bus licenses and invalidating every expansion card on the market."

The fact that IBM were selling mechanical boards into the 1990s was therefore not that relevant; Northgate ended up as a brand defined by their keyboards more than their PCs, and Dell was pushing mechanical boards past the debut of Windows 95.

I'd argue that a major factor was likely the growth of the white-box shop. They weren't tied into any long-term contract saying "we're going to build a million AT101s/Model Ms/etc", and there was cutthroat price competition. These were the sort of shops that used tissue-paper thin cases, power supplies where the CE certification just said "Nope", and PCChips motherboards with fake cache, but they'd sell you a desktop with the same "big numbers" (MHz and MB) for several hundred dollars less than IBM, Micron, HP, or Dell. They're going to grab whatever OEM supplier offers the best deal today, so long as it meets some minimum level of "not too weird or defective that people won't return it in droves."

There's probably also some interesting trends in the market for replacement keyboards at this time. As PCs became a home consumer thing, it wasn't "call IT and they'll get a like replacement on a service contract" anymore. It was "I spilled a soda into my keyboard, I'm going to go to CompUSA and buy whatever is cheapest so I can finish my game of Doom." You're dealing with a price-conscious consumer who's buying it seperately, so it's not like you can finance the costs of a better keyboard in the margin of a big system or service-contract purchase. I can recall buying rubber dome keyboards at the local Fry's Electronics for like USD10 each in the mid-late 1990s, and then coming back next month when I had worn them to death.

Now, the problem is if you start shifting mass production into the 'cheapest available' rubber dome board, it makes the mechanical products more of a niche-- tooling gets retired and assembly lines repurposed. So even when you have a brand that wants to source better boards as a competitive advantage, it becomes harder to even find them, and lost scale makes them even more expensive.

I think we can see some evidence for this in the presence of obviously transitional products-- the migration from full mechanical switches, to slider over dome with MX or Alps mount, to single piece keycap-on-dome designs. Using a slider in the short term means you don't have to retool your entire assembly process and find a new source of keycaps.

TBH, I could see that being a factor in the rate of evolution too. As long as mass manufacturers were using double-shot keycaps from firms like Tai-Hao, they sort of remained stuck with keycap-on-slider designs, which may have limited the cost-cutting appeal of going rubber dome. This could have bought a year or two. When laser and printed lettering became mainstream, suddenly you could use a full custom cap-over-membrane design for maximum cheapness.

Aside: you might want to write to the Red Hill people (https://redhill.net.au/ig.html). They seemed to have an interesting perspective on the industry (being one of those "white box shops"). Would love to see them do a series on keyboards like their CPU and hard drive pages.

Findecanor

23 May 2021, 12:33

Hak Foo wrote:
23 May 2021, 10:49
You're dealing with a price-conscious consumer who's buying it seperately, so it's not like you can finance the costs of a better keyboard in the margin of a big system or service-contract purchase.
Yes, and the consumers at home who had got a PC to replace the ageing "home computer" was not the most powerful driving force here.
As PCs got cheaper, more PCs got out into small and medium-sized businesses and into schools, that had a tighter budget.

I recall my "high-school" (equiv) in 1993 had mostly "white-box" 386 and 486 PCs, and a couple Compaq 386 and 286. The only IBM machines were a big file-server in the back room, and an older PC that had been relegated to print server duty.
Last edited by Findecanor on 23 May 2021, 14:22, edited 1 time in total.

esr

23 May 2021, 14:07

Hak Foo wrote:
23 May 2021, 10:49
I feel like IBM lost control of the industry around 1987. First, the Deskpro 386 hit, so they were now followers rather than leaders for the bleeding edge platform-defining aspects. Then, the PS/2 platform landed with a bit of a thud: nice machines, sold well enough, but it forced the rest of the industry to say "we're not going to copy everything IBM does, may as well follow the people who put a new CPU in without demanding bus licenses and invalidating every expansion card on the market."
Can confirm. I was watching that development closely at the time, because I had high hopes for the new 386 machines as affordable Unix systems. The PS/2 was definitely a damp firecracker, and that is about the point that IBM lost its grip.

But the keyboard transition couldn't have been that soon. If cost pressure were already making buckling springs a losing proposition in 1987, there is no way IBM would have signed its agreement with Lexmark in 1991. Four years was an eon in those days.

User avatar
fohat
Elder Messenger

23 May 2021, 14:24

esr wrote:
23 May 2021, 04:32

I feel like I might be reading parody**2 here - Anonymous Coward imitating a downshouter but subverting it in the last line.
You are probably right. This was posted in the ripster/r00tw0rm period and I would not be surprised if it was ripster himself.

User avatar
fohat
Elder Messenger

23 May 2021, 14:26

Hak Foo wrote:
23 May 2021, 10:49

maximum cheapness.
I love this term, I want to use it regularly. Thank you!

Post Reply

Return to “Keyboards”