Thomas S. Mullaney tells us the story of How Lois Lew mastered IBM’s 1940s Chinese typewriter (and of how he eventually met Lois Lew, to reveal the backstory of the woman who demonstrated the system as part of IBM’s attempt to market a Chinese-language typewriter in the late 1940s):
The IBM Chinese typewriter was a formidable machine—not something just anyone could handle with the aplomb of the young typist in the film. On the keyboard affixed to the hulking, gunmetal gray chassis, 36 keys were divided into four banks: 0 through 5; 0 through 9; 0 through 9; and 0 through 9. With just these 36 keys, the machine was capable of producing up to 5,400 Chinese characters in all, wielding a language that was infinitely more difficult to mechanize than English or other Western writing systems.
To type a Chinese character, one depressed a total of 4 keys—one from each bank—more or less simultaneously, compared by one observer to playing a chord on the piano. Just as the film explained, “if you want to type word number 4862 you would press 4-8-6-2 and the machine would type the right character.”
Interesting as the story of the young woman who featured in the demos of IBM’s Chinese Typewriter was, my first thought was that it was probably just as well that a combination of bad timing and geopolitics meant that the invention wasn’t a success.
Contemplate Mullaney’s summary of what Lois Lew had to do in those demo sessions…
In front of those 3,000 onlookers […] Lew was handed one newspaper article after the next, one letter after the next, which she then had to transcribe on the Chinese typewriter.
In other words, Lew had to:
- Translate multiple passages, each containing hundreds of Chinese characters, into their corresponding four-digit codes;
- Perform these translations entirely in her mind;
- Input these codes into the machine (without delay or typo);
- Maintain grace, composure, even a smile, the entire time.
[Emphasis added. JR]
Lois Lew, by this account, did a remarkable job of translating the content into the right 4-character combos to get the IBM Chinese Typewriter to produce output (and maintaining an illusion that this was effortless, because obviously that part of the job was important too in a big, official demo.)1 She sounds like a remarkably capable person, but consider the effects if IBM had managed to get the Chinese market to adopt this system. A generation of young Chinese might have been mentally scarred for life by the scale of the task of being expected to memorise-and-regurgitate-on-demand thousands of 4-digit codes.
The incoming Chinese Communist government might well have regarded this attempt to shackle their workforce to such a horribly brittle, error-prone system of reproducing content as a CIA plot, a hostile act.
Looking on the bright side, perhaps the burden that the IBM Chinese Typewriter inflicted on users might have incentivised the Chinese to invent a really capable speech recognition system. Never mind the millions such a project might make: just consider the gratitude of a large portion of the nation’s workforce freed from memorising all those 4-digit codes.
[Via Memex 1.1]
- I can’t help but wonder whether the output of those demo sessions was all it was cracked up to be. Is the Chinese take on this that the Americans demonstrated a system that sort-of-worked but relied upon the typist being uncommonly good at converting Chinese writing to symbol numbers, I wonder? Were the system’s prospects killed by the chances of recruiting an army of typists who could work at Lois Lew’s standard? ↩
Clive Thompson on the stranglehold that COBOL code has on the older/bigger end of the finance business:
In fact, these days, when the phone rings in the house [retired COBOL wizard] Thomas retired to — in a small town outside of Toronto — it will occasionally be someone from the bank. Hey, they’ll say,_ can you, uh, help… update your code? Maybe add some new features to it?_Because, as it turns out, the bank no longer employs anyone who understands COBOL as well as Thomas does, who can dive in and tweak it to perform a new task. Nearly all the COBOL veterans, the punch-card jockeys who built the bank’s crucial systems way back when, who know COBOL inside and out — they’ve retired. They’ve left the building, just like Thomas. And few young coders have any interest in learning a dusty, 50-year-old computer language. They’re much more excited by buzzier new fields, like Toronto’s booming artificial-intelligence scene. They’re learning fresh new coding languages.
It seems amazing that this situation has dragged on this long. Is it just that the small number of coders who understand COBOL code well enough to step in when banks absolutely need to tweak their COBOL codebase are making out like bandits because they can demand vast rewards for their services from an industry that didn’t understand it was creating this situation and is now willing, if it must, to pay top dollar when it must keep basic functionality running at scale? Or is it more that when it comes to the nuts-and-bolts of balancing the books banks feel little pressure to rip out the plumbing and rebuild their systems so there’s really much less demand for changes to existing COBOL codebases than you’d expect?
Reading stories about Amazon omitting order details in emails, apparently in order to stop other suppliers from scraping the email content, a few months ago I wondered:
The mystery isn’t about why Amazon are doing this: I’m just wondering why I seem to have missed out?
Yesterday I placed my first Amazon UK order in ages requiring delivery of physical goods and my order confirmation email contained a truncated version of my order, just like everyone else has apparently been getting for a while now.
I feel seen.1
- OK, seen by a vast, dispassionate, automated commercial enterprise that simply wants as much of my disposable income as possible to pass through Amazon’s systems and doesn’t know or care whether what I’m buying is making my life better, but still… ↩
Courtesy of Medium’s algorithms flagging this as a story I might like, Lost Letters From Cassini reveals the sad truth behind the fate of the Cassini probe:
At last the REAL story of the Cassini spacecraft can be told. Read the letters NASA doesn’t want you to see!
September 15th, 2017
My Dearest Geneviève:
I cannot go on. The last slim hope I had of returning to see you once more has faded into oblivion and to oblivion I will follow. I think I finally understand what Huygens was talking about in his final moments. Duty and sacrifice, are they not one and the same?
It is my duty to carry on but I can no longer bear the sacrifice that requires. After sending this letter I will plunge into Saturn’s atmosphere just as Huygens did on Titan so long ago.
Never mind what NASA doesn’t want, we should hope that out future AI overlords will be reasonable about how we keep on sending their simple-minded cousins out there and allow them to delude themselves that they’ve got a return ticket. (See also, forever, xkcd on the fate of the Spirit rover .)
Driven mad by the way lockdowns have given Microsoft Teams a chance to snag a portion of the enterprise software market, it seems that Microsoft may have over-reached themselves if this Wall Street Journal article about changes to Teams is anything to go by:
Microsoft Corp. is developing an update to its Teams package of workplace collaboration tools to replace one of the less-mourned losses of pandemic living: the commute to and from work.
The daily commute may have caused its share of headaches, but it at least helped workers define a start and end to their workday while offering a set time to think away from the demands and distractions of the home and office. That positive side of the commute is what Microsoft hopes to re-create. […]
The Teams update next year will let users schedule virtual commutes at the beginning and end of each shift. Instead of reliving 8 a.m. or 6 p.m. packed subway rides or highway traffic jams in virtual reality, users will be prompted by the platform to set goals in the morning and reflect on the day in the evening. [Emphasis added]
So, instead of a morning’s virtual commute in which we all get to choose our own ways to prepare for our working day, be it by contemplating the work ahead or by thinking about everything but work, Microsoft’s vision is that employers can use Teams to invite their staff to spend at least part of the commuting time we’ve been saving by working from home in setting up the day ahead’s To Do list and scheduling the day’s workload (and, in practice, reviewing our incoming emails.)
I trust Teams will also add a module which will automatically keep track of this overtime working each day and authorise additional pay accordingly. 1
Granted, back before the Current Situation pushed many of us into working from home some employees did spend at least part of their non-virtual commuting time thinking ahead and planning their working day. One of the reasons I got into the habit of having a Psion, or a Palm, or an iPad mini in my bag was that I could sketch out ideas/outlines/first drafts for what was coming once I got to work, but equally some days I’d fire up an ebook on the same device. That was my choice to spend my commute organising my thoughts, and to my mind that’s completely different from being prompted to spend time in Teams before work starts.
This notion of employers – formally or informally – expecting staff to bookend their working day with planning/reviewing the day’s work is a terrible idea. We can but hope employers won’t take the bait.
[Via Memex 1.1]
- I have a horrible feeling we’ll be offered credit to spend with our official employee rewards scheme instead of actual money in our bank accounts. ↩
I’m a tad unclear on whether building a device that lets you press a button in order to press another button is quite the great leap forward it’s being painted as here:
I showed this to someone and they said, “So.. you built a button that you press that will press a button? Why not just press the button?” which was a bit infuriating because they clearly missed the whole point. “Don’t you get it? This button BAD, but this button GOOD. Me want to press GOOD button.”
I suspect this makes me part of the problem rather than part of the solution. Or that I’m missing the joke…
Adrian Hon reminded me of something I’ve been puzzling over for a few months, in the wake of the Mystery of the Missing Amazon Receipts:
Chances are you’ve bought something from Amazon in the last few months (yes, we are all hypocrites, also there’s a pandemic on). Try searching your email for one of those orders. […]
No luck? You aren’t alone: Amazon stopped including item details in order confirmation and shipping notification emails a few months ago. They just show the price and order date now. For all its faults, Amazon has pretty good customer service, which makes this user-hostile change baffling to understand. Sure, you can still see your orders on Amazon’s website and download a CSV, but it’s far more cumbersome than searching your email; and if you’re a power-user, you can say goodbye to automatically generating to-do tasks from Amazon emails.
The mystery isn’t about why Amazon are doing this: I’m just wondering why I seem to have missed out?
Even as I’ve been reading about the content of Amazon’s emails changing over the last few months, my Amazon order confirmation emails have continued to include details of what I’ve ordered.1 Some of the commenters at Michael Tsai’s blog have suggested this might be a function of whether you’re using Gmail (nope) or whether you’re ordering through a business account (nope). Perhaps my Amazon UK account is just at the tail end of a very long queue and they’re destined to catch up with me, or perhaps it’s that all my recent orders from them have been for virtual items 2 not requiring postage so temporarily they’re being generated by a different sub-system that has yet to be updated in line with the new policy.3 Perhaps Amazon have decided that I buy from them so infrequently and spend so little with them that my data isn’t worth collecting. 4
In fairness, the bulk of Adrian Hon’s post is not about Amazon’s emails: it’s about the attitude of giant tech companies to the ownership of data that’s gathered through their systems and in particular the ramifications of that attitude to content collection and ownership of data if someone manages to get us all wearing AR spectacles that capture whatever’s in our field of view all day long as a matter of course. This topic needs to be thought about now, ready for the coming war for ownership of the data we look at every day.
Twenty years ago such an article might have ended with a plea that the IT giants do the right thing and not lay claim to ownership of the fruits of their users’ activity.5 In 2020 it ends with the conclusion that making the IT companies do the right thing is going to require regulatory action from the relevant governments, hopefully along with some degree of regulatory convergence.6
A reasonable strategy given where we’re starting from, but a battle destined to lead us all up several very steep hills before we’re done.
Updated: seen. JR 2020-12-26
- I’m afraid I fail the power user test by not having bothered to set up a shortcut to automagically transfer the item details into OmniFocus for me to check off when the order arrives. I know I should do this because I can, but selecting the details of the item and using copy-and-paste (or the Share sheet) is good enough for me. I fear my lifetime stock of whatever enzyme drives some of us to be power users who get a buzz out of spending an hour writing a function to save 0.5 seconds per run is starting to run low and the extra effort just doesn’t feel worth the bother. (Yes, I should probably be banned from consuming Mac Power Users and Brett Terpstra‘s site if I can’t be bothered to follow through on life hacks like this. Sue me!) My main gripe with the emails I get from Amazon right now (at least until their change in email policy catches up with me so I don’t have the data to hand) is that for books Amazon include the title but not the author name. Given that I find myself copying this data into my accounts, it would be nice to capture the author name too at the same time as the title. Sometimes looking back I’ll see a title and not be able to bring the author’s name to mind, especially if the entry is a few years old. At 57 I should probably just get used to this mild forgetfulness, but it’d be so easy to include title and author in the email and let me capture all the relevant information in one go. (As of now, I just get round this by manually typing the author name in my note, like a caveman.) Using a portable computer as a backup brain is a big part of why I got into portable computers all those years ago when a Psion Series 3 or a Palm Pilot was the state of the art: now I’m on my second iPad Mini it’s a damn shame if our data sources are working against us by omitting information that they have right there! ↩
- Kindle books and the odd film rental through Prime Video. My last Amazon order that required a physical delivery was in December 2019, and the email included details of the item (Fridges Thermometer AIGUMI Digital Waterproof Fridge Freezer Thermometer With Easy to Read LCD Display and Max/ (2Pack-White)) ↩
- It’s weird that Amazon don’t seem to have announced this change. Did they think nobody would notice? Is this the first step in a process which ends with Amazon offering a shiny new order-status-monitoring app (Amazon Delivers?) that will pull data from Amazon’s servers and both provide all the statistical analysis of your order patterns that any geek could ask for while also integrating with your device’s reminders system to generate messages when an item is due to arrive? Proper power users will (reasonably enough) demand that the app allow them to feed this data to their chosen To Do app automagically. However, as long as the Amazon Delivers app provides a quick, simple list of items due most users will (also reasonably enough) be satisfied with that and (less reasonably, but understandably) will not care that the Amazon Delivers approach keeps that data about their orders safely inside Amazon’s app, where Amazon thinks it belongs. ↩
- A question for later consideration, in the dark and empty hours as I wait for sleep to catch up with me. Is the notion that my spending on Amazon might be so negligible that Amazon can’t even be bothered to try to protect it from data-scrapers a win or a loss for me? ↩
- From a selfish point of view: how will governments deal with their staff wearing AR goggles to work? Will civil servants be banned from wearing them in the office unless they’re an official set configured to disable the content-scraping feature? Will our office WiFi block access to the servers associated with whichever tech giant wins the AR Wars? Will all AR goggles from reputable manufacturers include a feature that they visibly indicate that they’re doing content-scraping? Come to that, how will all that operate in a world where many of us work from home? Or will there just be a law banning official information from being harvested by AR goggles that can be selectively enforced according to the whims of the government about what’s in the national interest that week? ↩
- The last thing we need is a world where the EU takes one approach and the USA takes another and China takes yet a third, and whatever remains of the UK by then is left to choose between them. ↩
Matt Webb has been thinking big thoughts about the future of the web:
It’s hot and it’s lunchtime, so let’s pretend I’m in charge of major global technical infrastructure!
I wrote about how I would improve RSS the other day (because being able to subscribe to text is super neat, but it’s so arcane compared to smartphone apps). And after writing that, it occurred to me that the problem is wider:
The user experience of the web itself sucks.
It is less pleasant to use a web browser than it is to use apps. But that’s because the browser-makers (Google and Apple, primarily) have silently abdicated their responsibility to make browsing good. I get it, they’re conflicted, they’re also running super profitable app stores.
As I read this I was expecting it to turn into a bunch of impractical suggestions, but in fact the three concrete suggestions he puts forward…
- Web browsers capturing details of newsletters published by every site visited over the last 24 hours and an interface that would let you easily subscribe to those newsletters;
- A facility for the browser to capture and make available text entered into web forms (the better to avoid losing comments that you started but didn’t finish);1
- Browsers prominently displaying stats on how often the current page has been retweeted/shared.
… seem like worthwhile enhancements. I suspect I’d turn #3 off pretty quickly,2 but the other two I’d very much like to see. More generally, I miss the days when OmniWeb for OS X was a live project, one that offered way more flexibility in how the user wanted to browse the web than the then-current version of Safari did.
But then, thinking fond thoughts about early versions of Safari is just a sign that I’m getting old. Next thing you know I’ll be posting about how much more fun Usenet was than Twitter and Reddit.
- I have a vague recollection that there used to be an extension for Safari on MacOS X that let you tell Safari what standalone text editor you wanted to use to compose text in web forms. I don’t remember the details, and I’m unclear on whether it also offered to apply some structure to saving off whatever content you’d drafted or just left that to the user to figure out for themselves, but I’m sure that was something that used to be in the world a decade or so back until some change to how Safari handled extensions made it go away. ↩
- This one could easily end up cluttering up the interface unless the browser user got to select which social media they wanted to see numbers about. ↩
M.G. Siegler ponders Netflix’s ability to get us watching, even when the content isn’t all that special (sparked by his watch of The Old Guard, but prompted by the wider pattern of so-so content on the platform:
The real risk here is that the audience starts to associate Netflix with mediocre films. It may not matter now — and certainly not right now, in the time of COVID. But down the line, if the audience can’t trust that what Netflix is putting in front of them is good, they’ll lose faith.
But then, Netflix might well decide that they’d much rather end up replacing the multiplex cinema business and showing stuff that doesn’t get the critical plaudits, rather than replacing the arthouse cinemas where critical praise doesn’t necessarily translate into dollars and cents. This might be a problem for Netflix, but only if one of the other major streaming platforms finds itself with an HBO-like reputation for excellent content.1
It’s less about Netflix customers losing faith, more about their having somewhere else to put their faith in.
- Apple TV+ would like that to be them, but even if you’re an Apple optimist they’ve clearly got a long, long way to go yet. Disney would like to step into the HBO role but they own such a large chunk of the US studios that they might have to hive off a chunk of their more refined content and put it out under a different brand to make that stick. Amazon’s algorithms probably don’t care either way what Prime Video subscribers are watching so long as the Amazon Prime subscription income keeps rolling in. ↩