This, like almost every typing training system out there, suffers from my pet peeve: if you type an extra character or miss a character, then you end up with a series of “mistakes” instead of just one.
Why don’t these programs use some form of edit distance? I’m sure that one could come up with a way to freeze the edit beyond some distance from the caret to reduce the potential for spooky UI action at a distance.
I was trained to type two spaces after a period: it's in my muscle memory. This website treats it as a mistake. No typing teacher would downgrade for that. It's awful.
It's true that the space between sentences should be larger than the space between words, as it is in properly typeset text, though not twice as large. Using two spaces is a way to approximate that when you only have monospaced fonts. It's no longer encouraged in electronic text when using proportional fonts because the rendering software _should_ adjust the spacing itself.
It "should" adjust, but in practice most software do not render spaces after periods properly. There is no in band signal to distinguish periods for sentence ends vs e.g. Mr. Thus there is a clear argument for double spacing in all contexts
Buddy, there are rules and there are recommendations. The fact that it was common practice to use two spaces after a period up until the early 2000s in many situations shows that it is not in fact a rule, which is linguistic (as in the capitalization rules for proper nouns), but a recommendation, which is stylistic.
Stylistic recommendations can also be taken as rules, but in limited situations, such as a place of work saying we follow these guidelines (a manual of recommendations), which then become rules for that workplace.
Finally this link https://en.wikipedia.org/wiki/Sentence_spacing shows that their complaint has some merit, because the spacing used are not rules of the language, but conventions or recommendations imposed to improve the production of text based on the requirements and limitations of a particular technology.
on edit: probably shouldn't describe capitalization as being a linguistic rule, as it is an orthographic rule...heat of the moment, sorry. One might argue that the same applies to spacing, but I suppose that capitalization in orthography has been established for so long, and in so many different systems that it can be claimed a rule whereas spacing is still just a recommendation.
It was on typewriters because the fonts were monospace so the period would have some extra space to its left, the double space after it made it easier to see at a glance where a new sentence occurs.
I don't really get this. This is how it should work, no? The point is to type the word correctly; if you miss a letter or mistype, and keep typing, then yeah, it's several mistakes in a row?
Say I type out the Iliad absolutely perfect, but on the first character I accidentally type an extra 'R'. Would you say that I typed it 0% correctly? Because this, and many other typing testers would score it that way, simply because every character after that is now shifted over by one. Ironically, later on missing a character would actually improve your score.
No, the point is to type the word a you'd do in real life, that is, make the mistake, finish word/sentence/thought and then correct. And the correction would be changing just the first mistake
> the point is to type the word a you'd do in real life,
I don't really know about that. I think the point is to practice typing, and that may or may not reflect how you type in everyday life. For example, most people don't spend their time typing random words without numbers or punctuation :)
In any case, are you saying that if you correct a word during one of these tests, it simply should not count as a mistake? Because if not, then I don't see what the problem is again - typ.ing and most other sites note corrections, and distinguish between corrected and uncorrected mistakes.
If so, I guess we might just disagree on what the point of these exercises is.
I don't see how this can be so difficult to understand. If the second word in my first sentence were misspelled, does that make every other word or letter that comes after wrong?
My muscle memory usually presses ctrl-backspace when I mistype a word, which typ.ing actually handles better than most typing software, which is to say ctrl-backspace does what I expect, it deletes the word and I can retype it, which is what I do normally.
As noted in the default test, this site is by ZSA, a keyboard company. Their boards are pricey, but excellent. I have both their Moonlander, which sits on my desk at home, and their Voyager, which I went with quieter key switches (brown rather than blue) for taking on the go. Their boards are very comfortable; I highly recommend considering split, columnar keyboards like what they offer if you spend most of your day at a computer.
As a side, I switched to Dvorak when I first bought my Moonlander about 6 years ago. I have only used split keyboards with Dvorak, and I have only used traditional staggered keyboards with QWERTY. My muscle memory is well isolated between the two, and I have no trouble switching between them (though as it has now been many years since I regularly used a physical QWERTY keyboard, my WPM on one is definitely lower than it used to be).
I went with quieter key switches (brown rather than blue)
If you like tactile switches, try the Golden Kiwi from MoErgo [1] (also Choc v1, so fits the Voyager). It pulled of the near-impossible combination of a profound tactile bump, with only 42gf tactile force, while being much smoother and silent than Browns or Sunsets.
Only disclaimer I should add is that I was one of the beta testers for the switch. But I can't go back to anything else.
lowprokb also has some great custom choc switches; I found I ended up preferring the very lightest silent linear ones for all but the home row, for which I have a slightly heavier switch.
I second the recommendation. I use the Voyager, and it's a delight! I switched when I was suffering from a lot of wrist pain, and it's never come back. I only wish they offered a wireless version for a cleaner desk.
I also picked up and heavily customized a minimal (3x5+3 or 3x4+2) columnar split and a novel keyboard layout during Lockdown, and now can type faster with it than I could originally with greater comfort and accuracy. The high number of rolls actually felt pleasurable to type on compared to the scattered movements of QWERTY; at higher speeds, though, it's more demanding as the movements are so minimal and the sequences are more intricate. On long sessions, I can feel some strain on the tendons on the back of my hands; a keyboard with wells or something to grip like a Twiddler could help. Over time, I expect my speed to continue to advance as I sync my keystrokes faster; there is just substantially less finger travel from home row.
I also retained my ability and speed typing QWERTY by continuing to use it on a standard keyboard the majority of the time. Really, I just kept practicing the alt layout by typing out articles and books on the ortho and a laptop. At this point I think I'll maintain muscle memory for vi/emacs in QWERTY and switch on the fly to the alt layout for long entries.
I use Miryoku style meta keys, and layers for arrows, home-page(up/down)-end, mouse-keys, scroll-wheel, numpad, symbols, media, and F-keys. I'm thinking of adding combos for international letters and symbols next. backspace-delete and enter-space+tab are under thumbs like a Kinesis. Escape is a left hand combo. I also set a lock function for the layers with tap+hold and release with a designated key -- it works great with numpad, scroll, mouse, and arrows. I'd like to add a rotory dial.
For switches, I enjoy the light action Kailh box whites that came with my split, but I'd like to try out some silent low-pro ones for travel and more comfort. I have a second ortho with a number row and much heavier linear ones that I'm getting used to; so far it works great and I have more planned for it.
On a regular keyboard I've found mapping the capslock key to be a second backspace is much more ergonomic. I use it in conjunction with Ctrl to delete the previous word and then retype.
> I have only used split keyboards with Dvorak, and I have only used traditional staggered keyboards with QWERTY. My muscle memory is well isolated between the two, and I have no trouble switching between them
Mostly similar for me, but I never considered that the fact that I’m able to pretty comfortably switch between using my laptop keyboard which is QWERTY, and my external USB keyboard which is Dvorak based, even when not having used one or the other for quite some time might actually be helped by as you say that they are two different kinds of keyboards with different feel.
I've had dozens of keyboards in my PC life, so far. The earliest examples were Chicony clicky-mechanical boards in the late nineteen-eighties. Those were pretty cool (and they're very much NLA).
I've had innumerable membrane keyboards. Some were more useful than others. I had a very long and pleasant run of membrane boards made by NMB, which were good while they lasted.
Over the past 5 years or so, I've been mostly using boards with Cherry MX Blue switches -- assembled by whomever. Those worked fine, too.
And lately, in the past month or two, I replaced the one at home with a keyboard that uses Hall effect sensors for each key in place of any kind of conventional electrical switch. It's been fine; in fact, it's been quite lovely so far. It's a joy to type with.
---
I haven't named any keyboard makers here except for 1980s-era Chicony and PS/2 era NMB. Some of my keyboards have been good, some have been great, and some others have been trash.
That's my background.
What's yours?
What makes ZSA deserve "so much love"? Can you elaborate?
- Generally high quality boards. I expect to be using my ZSA boards for decades.
- Uses open, standard, programmable firmware (qmk).
- They have fostered a community around their boards (importantly making it easy to share and discover layouts and promoting 3D printable accessories).
- I have found their live support team to be very accessible and knowledgable.
It's hard not to be biased having spent >$700 on keyboards from them, feeling the need to justify my purchases a bit. At their prices it's easy to find high quality boards elsewhere or build your own board on open tech, but I think you'll be hard pressed to find similar offerings both with such excellent support from the company and such a vibrant community.
So straight-and-narrow columns of keys are what they're into, and that's a big push that is an important part of what they're doing?
(I did look at their website earlier and the details of what I found seemed to get wrapped up in the tiny details of some low-profile keyswitch or other. It's also possible that I am unobservant, stupid, or otherwise inept and that I missed the main concept, but that's what I found and I didn't think it was very interesting.)
I love my ZSA Voyager, but after getting into the split keyboard space and having wrist/finger issues, I ended up needing to go with something that supported a concave well, and built in/better tenting.
I've been using my Glove 80 ever since. But the Voyager was a nice step and I still keep it on my desk.
That does look very nice. My mentor at my first job was actually the one who got me interested in split keyboards. His was a Kineses Advantage 2, which is also concave.
The Glove80 looks close to what I want, but that thumb cluster looks basically as bad as the Ergodox one. Too many keys overall, especially pinky ones.
You don't really have to use them all. Realistically on each side I use ~3-4 of the thumb keys consistently and I have the others bound to just things I rarely do.
I also completely unbound the outermost row of keys on both sides. i.e. far left and far right. Since I found one of the biggest issues I had that aggravated my issues were stretching the pinky.
I also have a Kinesis Advantage 360 pro but I think its better suited for people with bigger hands than me. I do prefer its tenting setup though since its much quicker/easier to use. The Glove80's is a bit more custom but I think I'd have preferred if there was like a good mix of the two.
I have both of these keyboards (as well as the Moonlander, Voyager and Go60).
The thumb cluster on the Glove is significantly more comfortable for me than the Ergodox / Moonlander. I never found either of the latter boards particularly comfortable. That said, I only really use the three innermost thumbs on the Glove.
The Go for me has a reasonable thumb cluster but the middle “resting” key is slightly further than what I was used to. It took a few days to adjust.
If you want something really comfortable try a Svalboard, but it requires more training.
Maybe it's harder at that size where you have to move your hands, but I use a 3x5 split, and my eyes never leave the screen, so printed keycaps are just pointless.
Yes! I did this and love it. My layouts are heavily customized so no standard painted keycaps would do me any good anyway. I use a coherent set of key colors on various layers to provide reminders of what various key clusters are about. Looks great too.
Yes. Was my first ortho and I went full blanks and yeah didn't work out. Went back to normal keycaps to get to grips with it.
Now I'm on a Halcyon Elora but still use the Moonlander at home
Also a huge fan - I started off with a Moonlander and then upgraded (downgraded, number of keys-wise, I guess) to a Voyager. Takes a lot of getting used to.
If you care about the ergonomics of the thumb cluster, the Ergodox and others like it have a bad one. It sticks out too far. For a more modern and ergonomic thumb cluster, check out Pinky4, Iris, or Corne. I've been using a Pinky4 for years and all the thumb keys are easy to reach and comfortable. I haven't used the Ergodox EZ, but I did borrow an Infinity Ergodox for a few months once.
Thanks! I'll take a look at those. I'm coming from an Advantage 2 (that they discontinued a few months ago) and wasn't sure about the Advantage 360's increased price.. So I have been shopping around.
I accidentally did the same thing, and after 20+ years I've also found my QWERTY speed remains pretty good.
I can type QWERTY ok on a split keyboard, but I have to look at the keys.
I can't type Dvorak at all on a normal keyboard, but it's not a big problem. (It is annoying when using my laptop's built-in keyboard, but I probably end up doing that for the equivalent of all of about 2 working days per year, and even more rarely for any actual work.)
monkeytype.com is the gold standard here. Typing tests are a dime a dozen because you can make a bad one in 15 minutes, but Monkeytype seems to actually be made with care. Never tried typ.ing before but it seems better than the cheapo ones, but less polished than Monkeytype. It adds keyboard navigation and code snippet presets and syntax highlighting over Monkeytype, which may be why it’s posted here.
I loved it! As soon as I start to do these things, I get stressed and become much worse! Anyone remember the car game where you raced against others by typing? That was fun but forgotten the link.
I had some typing classes in middle school but they never stuck. The only reason I became a reasonably fast and confident typer—which ended up being amazing for everything from work (emails and Slack more than code!) to HN to writing a lot on Quora—was Type Racer :P
It's amazing how much easier gamification can make it for me to stick with something.
Graphite[0] is my keyboard layout of choice for anyone looking to switch. Huge fan! Had previously used QWERTY and Colemak. These types of sites help tremendously when learning a new layout.
The thought of learning a new keyboard is kind of scary. Is there any true difference? 30 years of typing on a QWERTY here. I wonder how long it would take to get to 75WPM on graphite. A decade?
>I wonder how long it would take to get to 75WPM on graphite. A decade?
Assuming you can already type that fast, 3-6 months at most I would think (with daily practice / full immersion). My best speed with QWERTY was 160wpm and I was able to get to 120wpm+ with Dvorak and later Workman in under a year each. One big tip I have is to not avoid typing particular things. Enable the punctuation, symbols, numbers. You're gonna be slow at everything at first, so might as well train in a well-rounded way. A friend of mine chose not to do this and struggles greatly with typing numbers and symbols. monkeytype is my favorite site to train speed, but there are some others that are useful for the first few days as you're drilling where the keys are and getting used to typing common bigrams.
Not nearly a decade. There are great resources that start you with just a few keys and add more over time. Once you have all 26 letters, it’s time to move on to MonkeyType or an equivalent. You will absolutely have reduced output for a few weeks. I notice a huge difference compared with QWERTY when typing on a MacBook all day. You should not switch layouts for speed, but rather for comfort. The GitHub link contains a few interesting graphics about efficiency if you’re interested, but the very high-level result is sizable comfort gains.
Probably around one month. Using Colemak Mod-DH on an external keyboard and QWERTY on my Macbook. My brain is switching automatically, cannot use them in a reverse order.
I also use Graphite and previously learned colemak-dh. I love it. I wouldn’t necessarily recommend it to others because it’s harder to install on Windows and Linux. Also, the symbol placement takes a while to get used to. The best time to switch if you’re courageous enough is after obtaining a split keyboard and during the holidays when you’ll have the spare time to learn a new layout.
I do love the fact that I can use this interface without touching the mouse or touchpad. Between the /, Enter and Esc keys, I can pretty much get to anywhere within the application. Even within settings, you can just start typing "Line..." and it will take you to Line Width setting. Wish more applications did this. Easier to develop, easier on the resources, and easier to automate. Alas, it doesn't help maximize engagement metrics, so most consumer applications never do.
I enjoyed playing with this. I should do some typing tutor on my new keyboard so that I can get faster at using it.
I think a typing tutor is better if it can re-sync when you have an insertion or deletion error. That is, if you miss one letter but everything else is correct, or if you mash an extra letter but everything else is correct, that is one error, not a full list of errors. It is distracting to have to look at your output to make sure you didn't make one of these mistakes.
It does some, uh, interesting things with the Dark Reader extension, though, at least with the default light theme darkened by Dark Reader. Errors are displayed. Corrections are highlighted. But there's no cursor shown, and no progress displayed for completed input.
It felt very strange typing like that. I somehow managed to get 65 wpm in that mode, but it was weird the whole time -- like my eye/brain combo had to focus on only one word at a time, without any lookahead, since I was typing into the void. It felt arduous in ways that are otherwise difficult to describe.
Switching to the site's dark theme and turning off Dark Reader seemed so much better, and was much faster. I got a fairly-lazy 88 wpm on the first go with it running with its intended level of feedback.
But the interesting part is that I was able to notice that my eyes wander all over the place while typing. I found that I looking several words ahead, and scanning the context, and watching the progress, and that my eyes don't move linearly at all.
I suppose that's a thing that I've been doing for decades. But I definitely never noticed it until Dark Reader made this website unintentionally run on hard mode and thereby provided a contrasting experience.
Weird stuff. I'm reluctant to call it a bug; instead, it seems like maybe it is a feature that could be further developed. It might help to train for accurately entering long arbitrary phrases (like passphrases, license keys, and credit card numbers) where visual feedback is deliberately limited.
Today’s typ.ing:
Speed: 127wpm
Accuracy: 100.00%
Training Set: English, most common 100 words
Capitalization: Numbers: Punctuation:
Lifetime Characters: 10982
I would not object to seeing my statistics compared to the statistics of others, what percentile I am in, what percentile I am in based on referrer source and operating system.
I wouldn't say that I'm up there in age, as I'm only 33, but I do think that I typed away faster when I was about 18 to 19 and gaming every day. But that the way that I typed led to issues with my wrists and hands later on.
I never had any dedicated typing classes like I know some younger people get now (like my younger brothers), so I just learned by doing and while I could touch type I had a bastardized technique that had my hands flying around all over the keyboard.
Nowadays, I type around 20-30 WPM lower, going down from closer to 120 to 140 to closer to consistently 90 to 100, but my technique is completely different than the past.
Two things I'd love: to try out this keyboard before buying it (I keep repeating myself, but I'd love a retail keyboard trying out store in NYC - Micro Center isn't cutting it)
And two - does anyone know of a keyboard that has some-trackpad-like (or pure scrolling..?) device pretty much where the trackpad would be on a laptop?
I spend time working on my laptop, get used to scrolling / moving the mouse "under" the keyboard (ie closer to me) using the trackpad, and when I go back to my desk with my full sized keyboard, I see myself reaching for a phantom trackpad.
This is really cool! Although a little disappointed that it punished me for using an emdash instead of a dash when an emdash would've been syntactically correct :)
Slow is smooth, smooth is fast. The biggest hits to typing speed are usually typos or fixing typos. If you slow down and try harder to hit the right key the first time, I'd bet you can hit 20wpm at least.
96.5% accuracy at 85 WPM. Would've been better except that they had _one_ space after periods and I'm still used to two - the UI didn't really indicate it very well for me.
Why don’t these programs use some form of edit distance? I’m sure that one could come up with a way to freeze the edit beyond some distance from the caret to reduce the potential for spooky UI action at a distance.