Brilliant But Doomed: Technology’s Most Magnificent Failures

The short lives and sad fates of twenty genuinely bright ideas.

By  |  Monday, July 27, 2009 at 5:59 am

5. Apple HyperCard (1987)

Apple HyperCardWhat it was: Apple’s software–invented by Macintosh genius Bill Atkinson–that let non-programmers create interactive hypertext “stacks” that incorporated words, images, and links on “cards” to convey information.

Why it was brilliant: As a concept, hypertext had been around for more than twenty years before HyperCard debuted without ever really emerging from the research labs. HyperCard (which shipped with every Mac at first) gave hypertext creation a point-and-click interface that made the whole process both easy and fun; it was arguably the most revolutionary thing anyone did with the Mac since the arrival of the Mac itself.

Why it was doomed: Well, it wasn’t, necessarily–for awhile, HyperCard was both popular and widely imitated. But three years after HyperCard’s arrival, CERN researchers Tim Berners-Lee and Robert Cailliau proposed an idea for a networked hypertext system which they called WorldWideWeb. (Cailliau had been a big HyperCard fan.) Within a few years, the Web began to apply HyperCard-like organization to the entire planet’s information. Rather successfully so.

If Apple had grasped HyperCard’s potential, it might have turned it into something akin to the Web before there ever was a Web. As former Apple CEO John Sculley told Cnet in 2003:

We weren’t insightful enough to recognize that what we had inside of Hypercard, essentially, was everything that later was developed so successfully by Tim Berners-Lee with HTTP (Hypertext Transfer Protocol) and HTML (Hypertext Markup Language). We didn’t call it that. But essentially, we had all that hypertext, radio buttons and linking capability architected in the original Hypercard. In hindsight, I wish Apple had recognized that we had a huge opportunity to go take our user interface culture, and our know-how, and applied it to the Internet. I think we would have had a very different story for Apple during the 1990s. But that, of course, is hindsight.

Its legacy: The International HyperCard User Group continued to boost HyperCard for years after Apple lost interest; it appears to be mostly dormant, but its Yahoo Group still sees traffic. SuperCard, a longtime HyperCard clone/rival, is still around. But HyperCard’s greatest legacy is the Web itself–and quite a legacy it is.

6. Lotus Agenda (1988)

bbd-lotusagendaWhat it was: A groundbreaking DOS Personal Information Manager from Lotus Development, the software company whose 1-2-3 spreadsheet was one of the most popular applications of its era.

Why it was brilliant: Confession time–I never used Agenda, and include it in this list because smart people such as Agenda überfan James Fallows have spoken of it so reverentially for so long. Fallows has written that Agenda is hard to explain, and while he gives it a good try, its appeal remains somewhat elusive. But if nothing else, the program–which not only understood English commands like Fallows’ example of “Lunch meeting at noon every Tuesday” but let you construct your own rules from scratch–sounds interesting.

Why it was doomed: Powerful PIMs have never been surefire software hits–see Ecco Pro below–and Agenda, like many of them, had a reputation for being impenetrable unless you truly dedicated yourself to figuring it out. Still, Fallows has written that the package sold 200,000 copies; that doesn’t sound like a flop to me. Lotus, however, lost interest, never updating the program after 1992’s 2.0b or bringing out a Windows edition. That puts Agenda in good company: Other promising applications abandoned by Lotus include HAL, Improv, Magellan (see below), Manuscript, and Symphony.

Its legacy: Two decades later, Agenda junkies still wistfully rue the lack of anything similar on the market today. One such fan happened to be Mitch Kapor, the founder of Lotus; in 2006, he funded the development of Chandler, an open-source project to develop an Agenda-like application for Windows, OS X, and Linux. Chandler inspired a good book but hasn’t taken off. Maybe Agenda was not only ahead of its own time, but ours as well.

7. Lotus Magellan (1989)

Lotus MagellanWhat it was: Yup, another DOS-era program from Lotus. This one was a utility that indexed the contents of your entire hard drive, then let you do lightning-fast searches.

Why it was brilliant: In an age when DOS’s limitations made file management of any sort remarkably tedious, Magellan’s full-text indexing and sophisticated file-format filters were remarkable, period.

Why it was doomed: Like Agenda, Magellan doesn’t seem to have been an outright failure in the marketplace–James Fallows says it sold a half-million copies. Once again, the biggest issue may have been Lotus’s inability to focus on any product that wasn’t an utter blockbuster. And maybe lightning-fast searching of your entire hard drive didn’t seem quite as important in a day when most hard drives maxed out at 20 or 30MB.

Its legacy: The basic idea behind Magellan resurfaced in such search utilities as X1 (from a company founded by Bill Gross, the creator of Magellan). Then it was integrated directly into operating systems, in the form of OS X’s Spotlight and Windows Vista’s Instant Search. Which proved that Magellan was a fine idea all along–it just took most of us a while to figure that out. And gee, is it going too far to say that Google is nothing more than Magellan for the entire Internet?

8. The NeXT Cube (1989)

NeXT LogoWhat it was: The first workstation from the company of the same name, started by Steve Jobs after his ouster from another computer company he co-founded.

Why it was brilliant: At the time of its release, the $6500 cube really did look like the next major advance in personal computing after the Mac: It had a 17-inch megapixel screen, a 256MB recordable magneto-optical drive instead of a 1.44MB floppy drive, a sleek magnesium case, and an operating system whose user interface made the Mac OS look nearly as quint as DOS. Like the iPhone, it was a product that seemed to live up to the hype that Steve Jobs showered upon it.

NeXT CubeWhy it was doomed: The NeXT, like all products, turned out to be less than utterly perfect–for one thing, people discovered that a machine whose only removable storage consisted of $50 mageto-optical discs was impractical. But NeXT’s big problem was that the market it targeted–university types who used neither PCs nor Macs nor workstations from the likes of Sun–didn’t quite seem to exist. The company reinvented itself as a software outfit, then sold itself to Apple for $400 million in 1998, solving Apple’s dire need for a next-generation operating system (and eventually making Jobs the CEO–you know he’s back at Apple, right?).

Its legacy: Every modern Mac on the planet runs OS X, the operating system built upon NeXT’s foundation; the iPhone is also powered by an OS X variant. It all gives the NeXT cube’s software an exceptionally successful afterlife. (Apple sold twenty times more iPhone 3GS units in the phone’s first weekend than all workstations sold by NeXT in its history.)

Video sidelight: This demo of release 3 of the NeXTStep OS provides a rare opportunity to see Steve Jobs wearing a necktie.



Read more: 

54 Comments For This Post

  1. Pascal Cuoq Says:

    More on the difference engine:

  2. Josh Bernoff Says:

    At Forrester Research we actually study this stuff to learn what makes technology succeed or not. The easy ones are products for which there is no need. The tantalizing products, like the ones you have here and the Newton, just barely miss for some reason.

    The main reasons appear to be timing, cost, and poor usability. These problems are all solvable in the long term, which is why you see successful products that do what these did, but in a way people could accept.

    Incidentally I created a methodology for predicting the true market size for a product based on all this analysis. It was called TechnoPotential. But by the time most products are ready to launch, developers are too far along for any advice of this kind. Result: just like the products you mention here, TechPotential was a flop.

  3. Patrick Says:

    Great article. I’d only add one thing; Babbage may the father of the technological magnificent failure but Da Vinci in the Grand Father. 😉

  4. dave Says:

    I worked for RCA on Selectavision player production in Indianapolis and Bloomington Indiana.
    The discs used detection of a variation in capacitance to retrieve the video/audio from the disc – hence the name CED Capacitance Electronic Disc. It was not like audio LPs where vibration of the needle was directly converted into electric signals.
    There was player that could play both sides of the disc eliminating the need for turning the disc over. I don’t know if that ever made it to the public.

  5. Dan Knight Says:

    As a longtime Mac-head, I’d never even heard of Microsoft PhotoDraw – but 10 years earlier I was using Deneba UltraPaint, which combined bit-map and vector graphics in the same program.

  6. Andrew Edsor Says:

    The London Science Museum’s Babbage machine is on permanent display and regularly demonstrated. Perhaps timely to give a plug for his wife Ada Lovelace who wrote his programs. In recognition, the US DoD, called their operating system Ada.

    When mentioning the VHS and Betamax systems, don’t forget the contemporary Philips 2000 system which was superior to both in terms of video and sound quality and had tapes which lasted much longer. I know, because I owned one.

  7. Brian Prince Says:

    The Sega Dreamcast was actually released on 9/9/99 in the US, not in 1998 as stated in the article.

    In addition to all its neat technical features, it was also the easiest console to develop for up to that point. Sega made it very painless to jump right in and start making games, which stood in stark contrast to Sony’s “they’ll learn it because they’ll have to” approach with the PS2, which was (at least for the first year or two) a developer’s nightmare.

    Features like edge anti-aliasing, which the DC did by default, took half a year for PS2 developers to hack together.

    The little grey box really was ahead of its time.

  8. drew Says:

    I wish I could go back into time to 2000 for the peice on the Sega Dreamcast; it shows the DOW at 11,192, or 2,000 points above today’s close….

  9. Harry McCracken Says:

    @Brian: Sorry, I know that the Dreamcast was released in 1999 in the U.S.; I used the Japan year in the header, then discussed the U.S. release in the text. (Confusing, I know…)


  10. Simon Says:

    I remember when Sega announced they were stopping Dreamcast development. I was answering phones for them and we had kids calling in tears and parents calling in fits of rage.
    One mother described how she had bought the Mega CD (Sega CD in the US, IIRC), the 32X and the Saturn, and she was angry because buying a console was an investment. Well, if anyone had been qualified to say what a great investment Sega hardware was, it was her…
    A sad loss to gaming, though, as todays Sega aren’t half as exciting as they were when they were making their own hardware.

  11. Gregg Says:

    IIRC, you couldn’t even touch those RCA videodiscs — you had to insert the entire sleeve into the device, and it somehow pulled the disc out and loaded it.

    It was nothing like LaserDisc, which you could handle like a normal vinyl record. In fact, the marketing for LaserDisc implied that you could really manhandle those things, because the lasers could read thru your fingerprints or something. Ha!

    As a kid I spent hours at the A/V store in the mall lusting after that beautiful Pioneer LaserDisc player. Even then I knew that the RCA VideoDisc player was analog crap compared to the digital goodness of LaserDisc. 😛

  12. dave Says:

    @Greg, The video on the laserdiscs was analog – much like composite video. There wasn’t the processing power in 1970’s (or 80’s for that matter) to convert a digital stream into a video signal. The audio was digital.
    Another project I worked on for RCA was DirecTV. We developed the video/audio receiver for that system and the processing power was just available (in reasonable prices) to implement MPEG-2 for the video – and that was the early 90’s.

  13. Says:

    What a nice trip down memory lane. I remember most of these first hand. Thanks for the look back. I love your coverage of old, obsolete technology. I am old and it’s nice to know someone remembers and takes the time to write about old things.

  14. nick Says:

    I bought a psion series 5 off ebay in 2001 and found it to be a pretty solid PDA if all you needed was text processing and a spreadsheet. It was a nice alternative to palm keyboard options. I never had a problem with the screen contrast and it had a decent backlight too. The article doesn’t mention that it had a touchscreen too.

    I considered getting a series 7 instead at the time, which was pretty much a netbook before its time, but from what I read it had less functionality that the series 5.

  15. Bill Says:

    I never heard of Microsoft Photodraw, but in the early 90’s, I was using Deneba Canvas on a Mac. It’s main feature, I recall, was the ability to combine vector and bitmapped graphics. It was a great product for creating figures for scientific papers.
    I also fondly remember the Psion, but I was using the older Series 3 model. The keyboard wasn’t nearly as nice as the 5, but the software was amazing for the time.

  16. changingwinds Says:

    Great story on how fast technology changes and how what once seemed cool now gets snickered at. It seems like yesterday when I was recording all my albums to tape. Big mistake. Now young people want vinyl. Me? I just ripped 700 CDs to my laptop since I got fed up with trying to find space for them. Digitize me, Scotty!

  17. Raj Says:

    I am the proud owner of a working Replay TV, so there’s one more to add to the count.

  18. Steve Says:

    I STILL miss Lotus Magellan – its capabilities are not matched by any built-in search. It’s ability to flatten, span, and select directory structures for searching was priceless, as was its ability to compare directories and files and subsets of files within directories in a single operation. Magellan and Laplink software/cable made it a lot easier to move to a new PC. IIRC, Lotus killed Magellan and refused to license it to anyone else for further development.

    I never could figure out how to make Ecco Pro work for me, though.

  19. skydvr Says:

    Still love my (multiple) ReplayTVs. Wish they had come out with a box capable of recording HD though….

  20. bufungla Says:

    Good to see the Psion mentioned. I had a Revo, and it’s still the best packaging for a PDA with keyboard I’ve ever seen. Slightly smaller than the dimensions of an old punched card, it was small enough to fit in your pocket yet had a keyboard large enough to touch type on. I could actually use my fingers to type on the thing, at a speed almost as fast as a regular computer keyboard. I’d pay almost anything to get hold of a PDA phone with a similar construction, but now the emphasis is on making the things smaller than a business card.

  21. Freda Says:

    Andrew, Ada Lovelace wasn’t actually Babbage’s wife but a collaborator, but thanks for giving her a mention anyway. Many see her as the first ever programmer, and she took the Analytical Engine conceptually quite a bit further than Babbage’s original idea.

  22. Spot Cool Tech Says:

    What a great review! These specific products might have been failures but aspects of each of them survive in products that went on to have success.

  23. Perry Says:

    Great article! One of my doomed favorites was HP New Wave, an object-oriented shell that sat on top of Windows 2.0. Along with HP New Wave Mail, it brought a lot of features, like dragging and dropping of files, er *objects*, into each other, and a desktop on which you store your objects, long before Windows had them.

  24. BDD Says:

    The Dreamcast lives on in the homebrew community, and there are still commercial games being released by small independent outfits. I even wrote a game for the system using the Fenix scripting language, but due to lack of interest (it wasn’t a clone, an emulator, or a 3d game, which seem to be the only games people are interested in for the system these days), I never released it to the public. It’s the first recent game project I actually finished…

  25. Zale 3 Says:

    Poor Babbage must be thrashing in his grave. I can hear him saying something like “what would you Modern Chaps call a Dyson Sphere?!!”………

  26. Jim McMillen Says:

    Lotus Magellan came with a trial version, MGSample, that did not have that great search facility. But it did have great file manipulation capability–renaming, moving, deleting, etc, individual files or groups of files simultaneously–all far faster and easier than anything else I have available. In spite of it being a DOS program, I still use it almost daily. That is one reason I insisted that my new computer run Windows XP.

  27. darkpand Says:

    Well, in fact, before Gateway 2000 Destination (in 1995), there was another similar project, the Olivetti Envision. It wasn’t a whole living room like the Destination 🙂 , but a vcr-style multimedia computer with remote (infrared) keyboard an trackball, and tv-out. It was brilliant, i always wanted one.

  28. John Bartell Says:

    Anyone remember the Newton. Although it’s success was underwhelming, it positioned Apple for the iPhone and iPod. Without the Newton, perhaps Apple would not be so successful with these products today.

  29. DT Says:

    I bought a used ReplayTV 4xxx series four years ago from an auction selling off assets from an IT company that collapsed under massive fraud and embezzlement by the higher-ups (kind of Dot-Com 2.0-ish). Got a lifetime subscription with, and two remotes. Within a short time, I’d upgraded it from a 40GB hard drive to a 250GB one, and hooked it up through component cable for 480p.
    While its resolution is no longer cutting-edge, its interface still beats the TiVo hands down, as does its commercial-skip feature. The ability to share shows between ReplayTVs of the same model line either in the home or over the internet was also a groundbreaking feature at the time, as was setting it to record over the web. I’ll continue to use mine until the call of a HD TiVo becomes strong enough and cheap enough someday, or until the TV guide service (bought by DLNA, as I recall) collapses.

  30. Robert Says:

    I only gave up my ReplayTV a few weeks ago, and then only because it was time to switch to HD. Frankly, other than having two internal receivers (the ability to record two streams at once), in almost all respects the DirecTV HD DVR has a slower and less capable interface than my old Replay. The loss of automatic commercial skip is devastating.

    When DirecTV bought the Replay assets, they shut down the web site and killed the remote DVR programming service. DirecTV should be complimented for giving us the iPhone app to program their DVR, but it needs the ability to create extended recordings for sports events.

  31. Chris Says:

    Atari Lynx


  32. Alex Says:

    As one of the few remaining Lotus Domino advocates out there, they knew how to build a product, and the knew how to build an AMAZING product. They were offering the SharePoint reality 10 years before anyone even grasped SharePoint. What Lotus/IBM didn’t know was marketing, and the eventual acquisition by IBM was the death knell.

    I once read that IBM’s marketing department was so abysmal, if they tried to sell sushi, they’d open restaurants with signs that said “Raw Dead Fish for Sale”. And I believe that. Lotus built some amazing tools – WebSphere is just a revved-up version of the Domino engine, after all – but utterly failed in marketing them.

    I’m probably going to end my IT days as a Notes-to-Exchange consultant, but I really miss the promising tools Lotus and IBM’s Lotus division offered.

  33. daemonios Says:

    Great story, very entertaining! I just can't believe the Amiga isn't listed there. It had everything going for it in the early 90s and it suddenly died out… Ok, it didn't so much suddenly die out as suffered from prolongued disease and faded out of people's memory before dying 🙂

  34. Xyzzy Says:

    The Apple IIgs belongs right alongside it… It was impressive for its time, but after Wozniak left, Apple management (Jobs/Sculley) did all they could to sabotage it, including ignoring/misleading game & software developers, barely advertising it, and so forth.

  35. Brett Says:

    Various members of my family still own ReplayTV 5000 boxes (after upgrading from 3000’s), and while I no longer use mine due to Replay never putting out an HD capable unit (which I would still buy in a heartbeat today if one were offered), my parents are still using the 3 that they purchased, mostly because there *still* is no decent alternative available to them to be able to record a show on one cable/satellite box and stream it to another box, which they use all the time.

    AT&T U-Verse has been touting this ability for about a year now, but still relies on DSL technologies that are distance limited and cannot reach their house with sufficient signal strength from the closest telco switch (despite being in the middle of the city). Anyone know of another service, Satellite or otherwise, which can do this?

    Using Time Warner’s DVR is painful in comparison – no 30 second skip, no streaming, and worst of all, no ability to search for a show – none. In that respect TiVo does it best, I’ll grudgingly admit.

  36. Chris R. Says:

    Years ago when Handspring was still selling Visor PDAs, I partnered with a friend in a venture to make Springboard modules for it. In a way it was fortunate that Handspring eliminated the platform when it did, before we'd had a chance to sink more than a few months of development effort into it.

    What you say about Springboard is true, but I'd like to add what we felt to be a major factor in its demise. The springboard pinout was based in large part on the pins of the Motorola 68k processors used in the Visor PDAs — many of the wires went straight through. This tied Springboard, and thus the devices that used it, to a specific, and outdated, hardware platform. We knew this was a potential problem when we got started, but we didn't know we'd hit that brick wall so soon.

    The Springboard module market was just getting onto its feet with tiny companies developing and improving their products over the short time they were around. It was sad to see it all get wiped out so suddenly.

  37. Clytie Says:

    I desperately wanted a Newton, but couldn't afford it at the time (making do with a Casio organizer). I did buy a Handspring Visor Prism several years later, and it was a terrific device. The flexibility of the Springboard module was impressive: you could make your Visor into a mobile phone (much smaller and much easier to use than dedicated mobile phones of the day), a GPS and even (according to a nostalgic thread on MobileRead) a personal massager.

    This MobileRead thread asked "What was your first e-reading device?" Quite a few of us started with the Handspring Visor. 🙂

  38. Jim Says:

    Fascinating series. I enjoyed reading about products I didn't know as much as those that were familiar. I, too, would like to see recognition for Ada Lovelace. Her contributions to the programming of Babbage's machines drove his designs far further than he likely would have managed alone.

    One additional legacy for the Handspring Visor Springboard: the smartphone. Handspring introduced a Springboard phone module that was it's stepping stone to the Treo. To my knowledge, it was the first viable integration of cell phones & PDAs. My wife still has her old setup, but has transitioned through a series of later smartphones.

  39. Geoff Says:

    I was using Laserpaint on a Mac in '87 long before I ever saw illustrator or Photoshop. It had vector and bitmap functions combined and even colour, which the greyscale Mac displayed as gradients, you had to print a cibachrome to see it. Anyway, years ahead of adobe.

  40. eric javits handbags Says:

    this post is a great one thanks for sharing

  41. noukat 2011 Says:

    You do a great job here.really very nice blog.thanks man for sharing this post

  42. Dancer Glimmer Says:

    Amazing the speed at which technology changes!

  43. david Says:

    The original Zune was revolutionary. It pioneered the concept of music sharing and file sharing over mobile devices. Its only problem was that is came after that cursed iPod.

  44. Tom in Raleigh Says:

    Great stuff here! I remember Lotus Agenda, and was just getting good at it when Windows came along, then when Lotus stopped supporting it. Fallows also was a big fan of something sort of like Agenda called "Zoot," although it never really did what Agenda did. Now we sort of have things like Onenote, but they don't come close to the same functionality.

    The Babbage machine video is excellent–really helps illustrate a chapter in James Gleick's new book, The Information.

  45. Wiiiindy Says:

    I could be wrong and probably am, but I think ADA was/is a programming language, not an operating system.

  46. Karus Says:

    Great article and run back through history. One honorable mention could also be BeOs, the ill-fated OS that was originally designed to be sold to Apple to replace their ageing OS. Unable to agree on a sale price for Be Inc., Apple CEO Gil Amelio instead purchased NeXT and the rest is, well, modern history.

  47. NLP Says:

    Ah…the Think Pad. I actually had one of those. At least I never had an Edsel.

  48. Liana Burvine Says:

    Attractive portion of content. I simply stumbled upon your website and in accession capital to say that I acquire actually loved account your blog posts. Anyway I’ll be subscribing on your feeds and even I achievement you get right of entry to consistently quickly.

  49. wateree Says:

    Grow tired of Top 10 or whatever lists. Yet like a Tokyo earthquake, they intrigue the watcher.

  50. alan Says:

    I'd have to add the Sharp Zaurus ( to the list. Linux-based PDA/handheld computer with a keyboard built in and a brilliant display. SysAdmins loved them as they could remotely connect to their systems via ssh and do remote administration right from the Zaurus.

    Sharp dropped the ball on US marketing and never release a version with capabilities beyond built in Bluetooth, in a time where built in wireless was becoming the norm.

    Still an excellent device.

  51. Anonymous Says:

    Wow, I actually remember having a Handspring Visor when I was 10. The only module I had was for a Smart Media card, which usually contained a 16MB one. Now I have an Atrix 4G with an internal 9GB and an SD card for another 8GB. How times change.

  52. jibran Says:

    It has been seen for a couple of years that liver disease is spreading a lot.I think the main thing is the level of fats increases inside the body which will effect the lever. العاب اطفال

  53. Pimples Says:

    Great story, very entertaining! this post is a great one thanks for sharing

  54. Hvac System Says:

    What a great review! I like your article .It tell about video game. It was very interesting .Thanking you for sharing.

5 Trackbacks For This Post

  1. Bad Tech That Couldn’t (or Shouldn’t) Survive | Says:

    […] Technologizer has compiled a list of gadgets that were doomed from the moment they set foot on the scene. The list includes impressive technological feats that were often just a tad too late or early. Take, for instance, Charles Babbage’s Difference Engine. Designed in the 1840s, this five-ton behemoth is arguably the first modern computer… or it would have been if it were ever built. Despite grants from the British government, Babbage’s design turned out to be too expensive to produce. Or QUBE, an early interactive TV platform from Warner Communications (precursor to Time Warner) that offered pay-per-view, interactive polls, and shopping from your remote — way back in 1977. […]

  2. The Twelve Most Tarnished Brand Names in Tech | Technologizer Says:

    […] The Fakest iPods Ever! Mouse Trouble: 20 Weird Pointing Device Patents An Atari Retro-Unboxing Brilliant But Doomed: Technology's Most Magnificent Failures The Thirteen Greatest Error Messages of All Time The Thirteen Other Greatest Error Messages of […]

  3. The Twelve Most Tarnished Brands in Tech | Technologizer Says:

    […] The Fakest iPods Ever! Mouse Trouble: 20 Weird Pointing Device Patents An Atari Retro-Unboxing Brilliant But Doomed: Technology's Most Magnificent Failures The Thirteen Greatest Error Messages of All Time The Thirteen Other Greatest Error Messages of […]

  4. Jobs Calls Out Android Over Porn — Think Geek Australia Says:

    […] Brilliant But Doomed: Technology’s Most Magnificent Failures […]

  5. Tuesday – Tech Tossed Salad « Tech – for Everyone Says:

    […] Brilliant But Doomed: Technology’s Most Magnificent Failures “The short lives and sad fates of twenty genuinely bright ideas.” Read […]