I think the article misses that the HP 9100A was developed after HP acquired the patent of Olivetti Programma 101 and its italian r&d electronics laboratory. Programma 101 was the first personal computer ever made and the first to actually fit on a desktop. It was also the first to use the magnetic card that was able to read and save programs, the precursor of cassettes and floppy disks or the first memory format based on "magnetic paint". The patents and the laboratory sold to HP were fundamental to fuel the revolution of the following years and to make the HP 9100A. Italians won that race at the time, more than a decade before Gates and Jobs. I had the luck to become friend of one of the engineers of the machine, Gastone Garziera, his funniest quote when he finishes to describe his masterpiece is "Very very easy, as americans say".
That’s not at all true. The 9100A was a descendant of a prototype developed by Tom Osborne in 1964. There was a later lawsuit from Olivetti that certain features (especially the mag card) overlapped their 1965 P101 patents, which was settled for ~$900K. No insult to Olivetti’s great achievements but this was hardly cause and effect.
This is repeatedly the case: Xerox and the graphical computer environment, Kodak and the digital camera, and other exemples Aaron Brown has that I forget.
Often because "that's not the business we're in". Kodak was a photographic film company, not a photography company.
What is great about this story is the way that Hewlett, an engineer himself, was able to use his intuition to achieve product-market fit, you cannot imagine an MBA who came up through the finance department having his sort of insight, he would never had taken on that risk in 1971.
There is a modern reboot of this, the HP-35s, LCD display but the same curved case.
An engineer who lives across the street from the CEO and carpools with him. Not so common anymore, though I seem to recall that Java started with an engineer who played hockey with Scott McNealy.
The CEO having a consulting company do a market survey and the deciding "ah, whatever, I want one of these, let's go ahead".
I think especially the latter story seems typical for Hewlett, in that he was neither a by-the-book corporate drone, nor an utterly reckless damn-the-torpedoes type, but somebody who was willing to trust his instincts and take calculated risks at times.
The story continues with one later CEO who did enough low level stuff (even if together with a partner who was a genius exactly for these topics) to be able to understand the products on all levels, who as a boy cold called Bill Hewlett:
Note: Steve Jobs, employed as a "technician fixing up and tweaking circuit board designs" even sold as his the improvement of the hardware implementation of the Pong successor bringing Wozniak to "help" him:
Jobs started his career with working on the pure hardware. Later, he had many opportunities to learn about all hardware and software aspects of the projects he lead.
I have the HP-35s. I really like it, but the batteries die in around 2 years even if you don't use it. Other than that, it's a really fun, tactile calculator that I find myself pulling out for simple calculations.
I would have to respectfully disagree. I think this can come across as stereotypical, and it may play well to an audience of engineers, scientists and programmers.
Individuals themselves possess brilliance, there are good MBAs as much as there are bad MBAs and the same for STEM professionals.
Rather we should say Hewlett had exceptional insight at the time, vs. attributing it to his upbringing and then making a sweeping statement on MBAs coming through a finance department.
Good MBAs tend to be good at being MBAs, which is an orthogonal skillset to being disruptive inventors.
Tim Cook has an MBA. No one could complain about Apple's financial performance, but the company has definitely lost the "Let's launch a market-defining product every five years or so" momentum and gloss of the Jobs years.
It's an open question whether this would actually have harmed Apple's performance over time, or whether the market cap would be even bigger if Jobs were still alive. (IMO it would probably have led to a much broader and more inventive product range which would have triggered earlier antitrust action - but of course that's just speculation and there's no possible way to back it up.)
The iPhone was an exceptional invention for Apple. It’s not guaranteed that Steve Jobs would have another hit.
And many would say the M1 and Silicon Macs are a considerable step up, and so is the slow but steady March of wearables via Apple Watch and Air Pods.
I do agree with you that there is a higher chance of innovation with Steve at the helm, but again he is a generational talent so is it fair to paint the MBA Tim Cook in a bad light when he has been reasonably successful in not just business but also innovation.
My overall point is smart people are smart people and generational talents are generational talents. They can be MBAs or engineers or any other profession. Creative MBAs will be creative and dull engineer will be dull.
I loved my HP calculators. I used them in college in the 80s, and grad school in the early 90s. Mostly HP15, 41, and 42. My last 42 is sitting in a drawer nearby, with a relatively fresh battery, and it works well.
I recall writing "programs" that performed some of the calcs I needed in the physics curriculum. I also recall being amused that I could calculate 1/3 (or 1/7) and add it to itself until I got (nearly) 1. From there this led me to study what fixed and floating point were, and think about the implications for the code I was writing (precision, truncation/roundoff error, etc.) to do longer calcs.
This eventually led me to skipping a postdoc, and working in the high performance computing industry. Where I am today.
I have the hp 41 cv app on my iPhone and it's fantastic. I paid once many years ago and have never been asked to pay again. A great group of enthusiasts keep it going.
That said, I've had recent occasion to do some programming for celestial navigation on a handheld- and the TI is a lot easier to use for that.
I have an HP-41CX+ app. I think I was using a free or entry version and I mentioned it in a review so the author sent me an upgrade to the full CX+ version. It's pretty great although it's complete overkill for me. It obviously lacks the feel of the physical calc but mine died years ago.
There's a good HP-42 app as well although I'm just more used to the 41 so that's what I use.
I credit the HP-41CV I had in in high school for growing my interest in programming. Through college it was an HP-48G. My father-in-law recognized my love for these devices later by giving me his old HP-35, complete with manuals and a couple accessories.
One of the more interesting accessories was his special HP-35 hardcore looking security lock. You would securely bolt the base of the lock to your desk, place the HP-35 inside, and connect and lock the cover with a key. Precious device back then!
I had a Radio Shack scientific calculator about that same timeframe. It was stolen during a lab. I then bought a 32Sii. Just wanted to thank the thief, as I likely never would have been a lifelong RPN user.
I got a TI to go off to engineering school because HPs were still a real budget buster. As I recall, a scientific TI calculator was still about $200 in 1975 dollars.
But calculators were dropping quickly in price and a couple of years later I was able to pick up a discontinued HP-55 which I used for years until I picked up an HP-41CV with financial pac for going off to business school. (I probably had one of the horizontal format HP RPN calculators somewhere in there too.)
HP calculators were great. I’ve owned several of the postfix based ones. (I.e. one enters 12, 30, + instead of 12 + 30)
The article may not make it clear how many efforts were going on in that timeframe to manufacture calculators. The first electronic calculator I saw at MIT was a desktop system by Wang; it had a Nixie tube display, see [1,2,3].
The first handheld calculator I ever came into contact with was in 1971 when I saw a student with a Bowmar Brain.
One interesting thought that occurs to me is that a mid-1970s vintage HP-35 would still be a perfectly good calculator for a typical engineer. (Leaving aside battery life because of the LED display.) That's not true of a lot of electronics of the era, especially any digital electronics.
(Amps etc. would be another example if you stick to purely analog inputs, but that's not digital.)
Right about the time IBM came up with their PC-AT (16-bit 80286, 10MB hard disk, yes: MB) HP came out with their version, same price, same specs, name started with a "V".
Over the course of the following year, each and every hard disk in every PC-AT failed. IF and ONLY IF you had paid $200 extra for a service contract (about the price of the drive) then, WHEN it failed, you could pack up and drive your PC-AT to whichever city had the nearest IBM service office, and they would replace the disk while you waited. Sorry about your data.
The HPs came with disk drives that didn't fail.
Guess which company ended up with the better reputation -- for "excellent service".
It is said the failed drives (made by Conner, IIRC) became an artificial reef offshore of Boca Raton.
Most buyers of early scientific calculators bought them to replace their slide rule. Early TI calculators had model numbers of SR-xx, which stood for "slide rule".
My dad bought one as a student at Stanford at the time and he passed it down to me for use in high school >20 years later. An amazing piece of technology.
Related to this history is how calculators used CORDIC, COordinate Rotation DIgital Computer, to compute trigonometric functions in real time with minimal hardware.
CORDIC was originally developed by Jack Volder working at Convair corporation in 1959.
I suppose it would make the math of calculating profits a bit easier in a age with much less fancy calculating equipment. Especially if sales were distributed exponentially or normally with respect to some variable like time or location.
My favorite calculator app is Free42 (https://apps.apple.com/us/app/free42/id337692629) which is a (AFAICT) faithful HP-42S reimplementation. Besides the obvious functionality, I really love the haptic feedback (configurable) and the graphics. Recommended (it's also free).
It's nice and would probably be my goto except that I used an HP41 for so many years that's just what I'm used to. (And TBH I don't need either to add up or multiply numbers from time to time :-) )
The Swiss Micro calculators are tempting but I know I'd just be indulging in nostalgia. I have zero need for anything beyond my iPhone app and a couple of HP-1x series landscape format models.
I refactored it to Swift UI but Apple didn’t provide backwards comparability for old iOS, which I didn’t know until it’s too late. There’s nothing that special in Swift UI that forces it to be 13+, but that’s life
It is nice. I will say that PCalc has been around for a long time, and does RPN (along with a lot of other stuff). PCalc has been a fundamental tool for me for ages.
That said, I’ve never become comfortable with RPN.
That’s great to hear. It is interesting on RPN, I got used to it and normal entry feels foreign to me now. I guess you need to use it for a while to get used to it.
Thing is I only do basic calculations with RPN, I can see how it may be jarring for scientific calculations or longer multi form ones (although there are many who are very comfortable and prefer this mode of entry).
You need to get used to how you chain calculations but if you do it on a regular basis it becomes second nature. I imagine the additional display lines of the HP-42 make it easier than the one line displays of most of these calculators.
The changes started in the late 1990s. Some blame this exclusively on Carly Fiorina, but I want to hear from more people knowledgeable about the situation at HP during this time, since I heard from somewhere that the board was interested in moving away from the HP Way, and they hired Fiorina precisely to change HP.
From the 1970s until the late 1990s, HP's calculators were renowned for their build quality, similar to how IBM ThinkPads were renowned for their quality. However, in the late 1990s and early 2000s, some high-quality calculators such as the HP-32SII and HP-48GX were replaced by calculators like the HP-33S and HP-49G, respectively, that did not have the same build quality. This led to HP's older calculators being sold on eBay and other secondhand markets at high prices. (Thankfully HP's calculator quality would improve in the late 2000s with the HP-35S and HP-50g.)
I've heard similar things about HP's printers and personal computers; there is a discernible difference in quality between those built before the late 1990s and those built afterwards.
I think it's mostly change in markets. HP reputation (as IBM Thinkpad) came from serious devices for serious tasks. When you sell to the average consumer, a race to the bottom occurs. Lower prices -> larger potential base -> no reason to make it fancy -> no research -> relying on brand name only .. and then large scale economics make you spend more time and resources on the more milkable segment of your sales so the core company suffers too (but investors don't care).
> I think it's mostly change in markets. HP reputation (as IBM Thinkpad) came from serious devices for serious tasks. When you sell to the average consumer, a race to the bottom occurs.
I wonder if this can be avoided or mitigated by creating a subsidiary or sub-brand to differentiate between the high-end and high-volume. Something like Toyota/Lexus, VW/Audi, etc.
Maybe nowadays, after the last 4 decades, people learned some lessons. It's indeed very important to keep a core skillset, whenever the money cow dies.. if your smartest left for a better company your left with nothing.
I guess the counter-argument is to claim that the market for the original quality was going away. This does happen, too: the first-mover can essentially spend monopoly profits on making something which makes engineers happy, if they choose to. But as competitors catch up, customers start to vote with their wallets for the minimum usable quality. The early leader may choose to sell the same low-quality items and charge 20% extra for their name, but they can't choose to sell the high-quality items at 3x the price, if there are no buyers.
Seems like the typical strategy of arbitraging a brand’s reputation by cutting quality and therefore costs, but still charging the same price.
Again this is the MBA mindset, a company’s assets, of which the goodwill of its customers is one, are to be stripmined for personal gain, then they move onto the next company like locusts.
Not to give my age away, but I grew-up with a Ti-83, and I prefer it to any HP calculator I’ve tried. It’s possible that HP saw that entire cohorts of kids were growing up exposed only to Ti and Casio calculators with their own proprietary ways-of-doing-things - and commoditisation in general - and HP didn’t have a way to lock-in consumers the way that Apple did with ITMS when MP3 player competition peaked.
I also grew up with a TI-83, though in college I learned about HP calculators and RPN and ended up buying a HP-48S on eBay; those were the cheapest non-financial HP calculators I could find on eBay at the time. I love RPN, but I liked my TI-83's ease of use.
Another factor that could have led to the demise of HP calculators was Texas Instruments' partnerships with textbook companies. When I was in high school, many textbooks had TI-83 examples and exercises. My impression is that HP's calculators were marketed to college students and engineers, while TI marketed its calculators to the much more lucrative middle and high school market. Thus turned out to be a winning strategy for Texas Instruments.
As a computer scientist/software engineer, I seldom use handheld calculators anymore. I use the Unix dc command whenever I need to make quick calculations, and I have LibreOffice Calc, GNU Octane, and REPLs available (such as Python and Common Lisp) whenever I need to perform more complex calculations. If I must calculate something away from my computer, I can use my iPhone's calculator app. I would imagine that this killed the traditional market for HP calculators; why spend the money on a bespoke handheld calculator when most engineers have access to computers and smartphones?
However, in a classroom setting I wouldn't be allowed to use my iPhone. Texas Instruments did a good job courting the classroom and standardized test markets. HP might have figured that it didn't want to compete against TI in this market.
With that being said, I wish I didn't miss out on the HP-15c special edition that was on sale briefly in 2011; I didn't hear any announcements until it was already sold out.
I agree that it's hard to see how anyone gets $500 worth out of the HP-50G. I use i48, the HP-48 simulator for iOS. The only thing it lacks is that keyboard feel.
When I was in high school virtually everyone had a TI calculator but I couldn't afford one. At college we were required to own an HP-48. I don't recall paying $500 for it, but maybe I did and I just forgot. It doesn't seem like the kind of money I would have had at that time. I still have it and it still works perfectly, which puts it in league with my Tektronix analog oscilloscopes in terms of legendary durability.
Not to give my age away either, but in high school, for the rare classes that needed them, all we needed was a generic scientific calculator that included logarithmic and trigonometry functions.
In engineering school, the HP-28, -42 and the new -48 series calculators where dominant. The build quality and "feel" were great. I don't even know if there was an equivalent TI model at the time.
I would say the hp-Agilent split was really the turning point but the seeds were planted much earlier. Apparently the board and various advisors realized that the PC market was going to be huge and wanted to grow that part of the company. hp PCs were okay but splitting off the instruments, which were really the original hp business and focussing the new hp on PCs and consumer products was effectively what ended the original company. It wasn't until the Compaq acquisition (which also brought the old DEC into the fold) that the transformation to consumer was complete.
At that time injket was doing so obscenely well that a lot of mistakes were paid for and papered over from those profits.
I joined right around that time, a few months before Carly was hired as CEO. She was okay at first but got worse and when she left we were singing "ding dong the witch is dead", you could actually hear it throughout the entire floor I was working on. Unfortunately, the CEOs up to Meg were a total disaster, I think with Meg the recovery started and now, after spinning off HPE it's been a super-fun place to work at again.
Despite all the bad press there's actually a ton of interesting R&D going on still but the consumer doesn't really hear about it. It's a deal with the devil when you switch from selling to nerds to grumpy consumers around the world.
Yes. When they decided to give you the printer but sell the ink for $1000 per gallon.
This is what happens when you let accountants make business decisions. I'm sure it looked like a real golden opportunity at the time. But what it actually did was kill the home printer business --- or at least most of it. Now, most home printers aren't worth having.
