Will word processors start a home working revolution? (1979) [video]
youtube.com
youtube.com
Also, this was a window when the printers unions in the Uk were going on strike as the first journalist-direct-entry systems were coming into the news print business. Rupert Murdoch went full bore with a new clean build, shut down the old printing, and locked them out. Huge strikes, police on horseback, riots.
ICL was trying to sell the home office/desktop office concept, it was still very much about the styling of a giant desk/cab with a tiny monitor and built in printer, and the key issue was the quality of the keyboard (as ever)
It turns out that seeing more than one line of input in a buffer helps. Who knew?
One takes years to be an expert, the other weeks or months.
Same with a typewriter, vs a wordprocessor + printer. Imagine, typing entire documents fast, with no typos, whiteout, or correction capability.
That's expertise, that's learned skill.
No wonder a word processor, even in the early days of dot matrix or daisy wheel printing, was preferred.
We were expected to proofread the living bejesus out of our stuff from RUNOFF (the pre-troff but very similar word processor which was on a Dec-10 amongst other things) on lineprinter, before queueing up to print it out on daisywheel. The paper was very satisfying fanfold with increadibly fine edge strips with the sprocket feed, tearing it back into loose sheets before binding was great. Well.. right up until you spotted a typo and had to go back and adjust the markup.
Yet at school, the head office secretary had a daisy wheel, and we all coveted that crazy clean and perfect output.
Graphics output be damned, it was perfection.
So I can well imagine the resource constraint where you worked.
I was thinking about them the other day and got nostalgic for the tractor feed fan fold paper and dot matrix printers. Then I checked Amazon and saw how expensive they are now and the nostalgia faded quickly.
You could leave high school, do a typing test and enter the public service to join rooms full of people typing up ugly, handwritten garbage from the other public servants. The main impediment to doing that work wasn't the equipment, it was the quality of other peoples handwriting (or shorthand).
By the time the 1980s rolled around, you could get typewriters with an entire line of memory and even with erase capabilities but those were conveniences for less experienced typists. If you were typing as a vocation, you learned not to screw up enough that you didn't really need an entire page of text to get it right.
1. CA, central valley
In a few decades, we will be left with skilled jobs that only people with certain innate abilities can perform (after substantial training), and menial jobs that anyone can do with very little training.
This is not to say that I disagree with you, but being adaptable is more important than being just skilled. A lot of those typists eventually became secretaries and then spread further out to different parts of organisations, and a lot of them are still there. It’s only the people who don’t adapt who get left behind, and I’d like to point out here that I’m not a fan of that term. As I am Scandinavian that is probably of little surprise to people, but I think society has a responsibility to keep people useful so long that we require them to work until they are 73+.
For example I would never consider replacing a gas boiler myself as it stands, but what if it was just a slot in unit of standardised size with simple idiot proof connectors no one could get wrong? What if it's an electrical heat pump with no explosive gases or carbon monoxide safety to worry about? Even if the home owner doesn't want to replace it, it might get simple enough that a lightly trained delivery truck driver could do it. For example the delivery guy installed my washing machine, not a plumber.
I come from a DIY family, I don't think skill is the primary barrier. The "why would you do that when you can just pay someone to do it?" attitude is pervasive. That's not a dismissive characterization either, that's paraphrasing as best I can what a friend said when another friend and I were on the couch watching a youtube maker video. And it's an attitude I've heard again and again over the years.
The correlation between "almost any human can do it with sufficient training" and "amenable to automation" is weak to non-existing.
Supplemental income for a home bound person, carer role, that kind of thing? I might help.
So, it's marvelous marketing. It is cheaper over time for the French than bring out telephone directories every 2 years. You just plug in to the central computer, you find your numbers at the same time you can find all sorts of other things. You'll be using word processors and text processors as information units as links in an information chain, not just to type.
And when that happens, when you can get the technology into the home, when you can get the technology away from the office, when you have word processors and text processors talking to other text processors and that point you don't need the middleman or -woman, do you?
He is referring to the very successful story of Minitel[0](Videotex) which began to be implemented widely in 1982. It is relatively unknown outside of France but in France just part of their naturally shared "(3516-)" IT history; peaking in 2000 with around 25 million users (40% of the population). Xavier Niel, a french billionaire, made his first million by launching a "sex oriented chat service" colloquially known as Minitel rose in 1987 at the age of 19.
For [1] anyone who wants to dig deeper into that very peculiar french history.
[0]https://en.m.wikipedia.org/wiki/Minitel [1]https://m.youtube.com/watch?v=HOhK9bgQo8g
Times have definitely changed
The quote, which I'm not finding elsewhere online:
If we do it all wrong then it could be an absolute disaster. It's the biggest aid to totalitarianism you could ever come across, if you think about it, and that must be avoided at all costs. On the other hand it's the greatest boon to decentralisation and people fulfilling themselves.
-- Cy Endfield
Sherman has a book with Phillip Judkins:
Glimpses of heaven, visions of hell : virtual reality and its implications
London : Coronet, 1993.
Summary: The authors ask what virtual reality is and the effects it may have on our lives. They look at the technology, finance, designers, manufacturers and users; and at the consequences of VR in the workplace, in education, in military and medical establishments and on the wider shores of recreation.
https://www.worldcat.org/title/glimpses-of-heaven-visions-of...
https://archive.org/search.php?query=creator%3A%22Sherman%2C...
Alternatives to typing
I'm really intrigued by the one-hand typing device shown at the 8 min mark in the video. I'm impressed the inventor of the Microwriter wrote an entire book in the one-hand typing device. The narrator say it took him about 30 minutes to achieve competence in using the device.
Word processors (WP)
By the early 1990s (maybe earlier), GUI word processors were widespread, but the word processor does not seem to have changed much since then. The past 15+ years has been about implementing word processors in the browser (with fewer feaures than the more powerful 30+ older desktop word processors). Progress? Or re-inventing the WP again in the browser?
Use word processors to create more documents for people who do more knowledge work so that they can get better at doing it.
Sigh. There was a time when most of the work was not knowledge work. And we are sitting on the back of that right now. Literally *EVERYTHING* we do was the kind of work where people had to sit together or commute to get it done in groups each day.
"Mom's going to the office" meant something.
Sure most work is back-breaking or mind-numbing. But someone has to do it. And our parents did it. And because of them, we are entitled enough to say that we all deserve to work from home.
An example of this I read elsewhere - Alex Honnold climbed a mountain with no safety equipment in 4 hours when it took weeks and plenty of safety gear a century ago. What changed? It was knowledge that people shared that made them more productive. Not just climbing routes, but fitness and dietary knowledge. For example, Honnold does yoga to stay in shape, which few climbers were doing back then.
So when you say “real work” instead of “throwaway”, you’re underestimating how important knowledge is.
I think you're mixing up some things Honnold has talked about here.. modern science (regarding training and nutrition), climbing gear, and climbing analysis (techniques used to further the sport of climbing) have enabled modern climbers to climb things that were considered impossible even 20 years ago. A 5.9 (considered the hardest possible grade 40 years ago) is now considered fairly casual and can be free climbed (not the same as free soloed) by people with less than a year of training.
Freerider in Yosemite, Honnold's most famous free solo, didn't get a first free ascent until 1998.
The time required to climb these things may have gone down also as a result of better equipment, but free soloing is a very niche subset of climbing. It's always been much faster than protected climbing. Honnold's specific training enabled him to climb it without protection (the one time anyway), but it's not a thing even most professional climbers would aspire to. They're mostly focused on harder grades, though outdoor speed climbing has a bit of a scene too (but still involves some protection)
I specifically spoke about free soloing to take away the impact of better climbing gear. I implicitly ask - why couldn’t El Capitan be free soloed a century ago? Yeah sure, the shoes are slightly better but that’s about it. The main difference between Honnold free soloing in 2018 and a hypothetical climber free soloing in 1918 is knowledge. Honnold speaks a lot about the various routes through El Capitan mapped over the decades by others. Their freely shared knowledge is his productivity gain.
I agree with you, but a century ago, Freerider was outside of anyone's capabilities (shoes aside). My understanding is that Honnold has talked about just free climbing freerider being impossible a century ago as a result.
If people are capable of free climbing something, they're capable of free soloing it (whether it's wise to do that is a different story)
> The main difference between Honnold free soloing in 2018 and a hypothetical climber free soloing in 1918 is knowledge
What I'm saying is that "free soloing" isn't relevant here, free climbing it in any fashion would have been impossible
To be honest, if we're talking purely about the difficulty of the moves, "free soloing" is easier than protected climbing because you have less weight and don't have to get pumped out trying to manage gear and protection.
I'd argue that elite climbing is a post-scarcity activity. I suppose it always was... in a way, the question starts to become, how accessible can we make it, and to who?
Paradoxically both the "decolonizers" and the "imperialists" (e.g. capitalist maximisers) are likely to be unsatisfied with the idea that it's basically a status economy now; people share tricks voluntarily because they love the work and the cred of introducing something novel and sharing is worth more than having secrets. Climbing is bought and paid for in climbing reputation which is only very loosely related to the "real" economy.
When a broader range of humanity is introduced into reputation economies that can look like demand destruction since it suppresses demand for certain kinds of easily measured economic activity and replaces it with activities and value that are harder to measure (not primarily $ denominated).
[All this ends up looking like people getting pissed off because ideologically unaligned people benefit from climbing just getting easier, which in general terms IMHO should be looked at as a good thing because human capability is increasing. Yes there will still be a lot of hocking gear and sponsorships in the meantime, reality is messy].
At the workplace, it definitely depends on one's field. When I started at my job 25 years ago, people still word processed and printed a lot of documents. Today, most text is entered in the e-mail or chat editor. If someone wants to spin up a document, they pull up whatever editor is provided by the workplace, or Google Docs, but the problem of finding out how to use all of the formatting features is bypassed by just not using those features. Text has gotten more informal. Graphics are added by using a screen shot capture utility. Meanwhile, the techies are using Markdown, and even the word processors (or whatever online editors) recognize Markdown tags and automatically apply formatting.
Meanwhile, most word processors have crappy UX and require training to figure out the mystic incantations to get the presentation right, and then screw it up at save/export time. We could definitely do better.
Which is a long explanation for “your needs exceed the capabilities of that tool, and thus your dislike is rational.” The various extensions that add additional capabilities, such as the ones you describe always force me to consult documentation (which kind of defeats the purpose) and, as you note, they all suck balls.
A constant of my own career has been pushing tools well beyond their reasonable limits. That's usually an exceptionally reliable sign that I need to find a more appropriate tool.
Markdown is a tool for conveniently generating rudimentary HTML quickly and with low overhead. Using tools such as Pandoc, this extends ot other markup languages. But the point remains that once you've pushed Markdown to its limits, you should choose a more appropriate tool. I lean toward LaTeX myself, though HTML is often useful as well.
[0]: https://github.com/DaveJarvis/keenwrite
[1]: https://github.com/DaveJarvis/keenwrite-themes
[2]: https://github.com/DaveJarvis/keenwrite/blob/master/docs/scr...
Probably my main uses of Markdown are for Jupyter notebooks and services that render it automatically such as GitHub.
This comment was generated by GPT-3 given the following prompt:
- Writing as we know it may be a thing in of the past
- GPT-3 will automatically rewrite sentences into proper english
- People will only need to articulate ideas instead of worry about structure and grammar
Articulating and broadly structuring ideas is 99% of writing; comparatively very little time is spent refining grammar and low-level structure. (This is in contrast to programming, where high-level architecting and low-level implementing take roughly the same amount of time.)
As a sibling poster points out, simply concatenating your three prompt sentences with a couple connective words (and a couple grammar fixes) already yields a well-written paragraph:
“Writing as we know it may be a thing ~in~ of the past, [since] GPT-3 will automatically rewrite sentences into proper [E]nglish. [Thus,] people will only need to articulate ideas instead of worry[ing] about structure and grammar.”
I think this feature of word processors can be considered a sibling to markdown, not a child. They each seek to turn plain text conventions (easily typed and understood) into proper formatting.
And I just edited the text twice before sending. :)
Word can’t do layouts, so you gotta use some Adobe product for that, but some parts don’t let you copy in outside text, gotta retype it.
If you want to update a graph, instead of clicking on the chart to update the data you have to go through 3 or more programs to get a publication ready image.
Never mind there isn’t a git equivalent in the space.
On Linux, you could turn on autosave, then tie it in with a short script for inotify + auto checkin + update.
Very messy I guess, would probably be easier to write a patch for libreoffice, tied into undo/redo edit history.
Seems like a killer feature for some people. Infinite rollback.
Would work best in an optional separate window, so you could copy historic text back in.
Imagine a full file upload every change.
Instead, having a git style history for edits, right in OOo would be awesome. Maybe a ${filename}.hist file for each file.
You could exclude this from git/svn/dropbox whatever backups too, so only purposeful saves get sent upstream.
The GUI interface, its nuances would be key here.
Imagine being able to open the history window, scroll to, or search text of all changes, then click and be taken to the location in current document, with a placement option once there.
That's next gen history, and gp is right, why isn't this a thing?
https://support.microsoft.com/en-us/windows/file-history-in-...
MacOS also does the same via Time Machine.
https://support.apple.com/en-ca/guide/mac-help/mh15136/mac
Other options are Google Docs (the editor has version history support) and cloud storage services that support versioning (Google Drive, OneDrive, Dropbox, etc).
Nonetheless, Google Docs is a much better tool for most ordinary users than Git. With Google Apps Script you can easily use Google Docs as a CMS (content management system).
If indeed your wife actually needs to use Git, instead of teaching your wife markdown, make a bunch of customized headings and explain to her she must not use all of the formatting options in Google Docs.
Techies often tend to try to convert ordinary people to "techie technology." It is usually a mistake.
It's not nearly as good at merging complicated histories as git, and doesn't help you keep commits atomic when working with multiple files. But that's seldom much of an issue with human language texts.
Google Docs' suggestion feature is probably better than reviewing pull-requests for normal people. Its ability to let multiple people edit the document at the same time can be really great (but would be crazy to do when editing source code).
EDIT: Markdown on the other hand is far too simple to format formal writing. It was never the intention of Markdown to be that.
Here are some Markdown documents I've written and/or typeset using ConTeXt:
* https://impacts.to/downloads/lowres/impacts.pdf
* https://www.docdroid.net/VQU9VEZ/design-pdf
* https://www.docdroid.net/rHlVyPV/autonoma-pdf
* https://dave.autonoma.ca/blog/2020/04/28/typesetting-markdow...
My Typesetting Markdown series[0] does a deep dive into how to keep content and presentation separate, which is much easier using ConTeXt than LaTeX, IMO. Part 8 of the series delves into typesetting annotations. As for TeX Live, I agree. The instructions for installing ConTeXt on Windows is basically: download, unzip, run two batch files.[1]
That said, ConTeXt looks great for my purposes! So much more is done out-of-the-box, the documentation is well-laid out, and I can see looking at its main pages how I could set up formatting for a style guide myself, unlike LaTeX. Thank you.
Absolutely. I developed KeenWrite[0], a free and open source desktop editor, to replace those shell scripts with a graphical user interface and command-line interface[1]. There are several starter themes you can use, as well.[2]
[0]: https://github.com/DaveJarvis/keenwrite
[1]: https://github.com/DaveJarvis/keenwrite/blob/master/docs/cmd...
I'm seeing you'd submitted an eight-part series, "Typesetting Markdown", first part here:
https://news.ycombinator.com/item?id=20003913
That got no traction on HN, but I'd love to see that included as a 2nd-chance pool or submitted feature.
I recommend to be wary of changing someone else's workflow, however. Particularly if it is someone you live with. Unless the other wants to change their workflow, they won't accept it. And if something goes awry, you're the first person they call, and make responsible to resolve the issue NOW. For example, when stuff breaks a day before a deadline, you know what you'll be spending that day on :-)
Google Docs (or other online word processors) + revision history [0] works just fine as a distributed version control system for the vast majority of users.
Revision history, track changes, sharing, autosave backup, etc are all included into Word these days (with a corporate office sub anyway). And it does it's job.
However there is no standard tool like git in the publication space that everyone uses. There are dozens of competing tools, that all do a poor job of keeping a project together. Some important features being: tracking support documents, controlled sharing, revision history to multiple types of documents, bundled with an easy to use UX/UI.
Incidentally, git supports non-wysiwyg typesetting just fine.
In any case, almost no one apart from Knuth uses TeX anyway. At least not directly. Almost everyone uses some kind of macropackage like LaTeX.
For most people, markdown would probably be the saner input format than anything TeX derived?
(And I use and like LaTeX.)
> Incidentally, git supports non-wysiwyg typesetting just fine.
If you add custom diff and merge handlers, git should be able to support wysiwyg, too.
There's also LyX, which is sort-of like wysiwyg, but produces LaTeX output. Git would handle the produced LaTeX just fine, too.
Research Publications is more than just Academia. The de facto standard in industry is Microsoft Word. Everyone accepts word docs and expects to receive them. Sure, some publications take tex family documents, but not all. Why use a tool not everybody takes.
Git is too obtuse of a tool to get not programmers to use.
You also have people of various skill levels using these docs. From MD/PhDs and lawyers all the way down to secretaries with high school diplomas. The only ubiquitous application between them is Microsoft Word.
Only a small subset of the research world uses LaTeX.