Seth's Blog: Hardware is Sexy, But It's Software that Matters (typepad.com) 78
American author and entrepreneur Seth Godin argues that though hardware is nice and dandy, it is the software that matters. And not just software that runs on a computer, "but the metaphorical idea of rules and algorithms designed to solve problems and connect people," he writes. Godin has used the piece to note how Apple has increasingly grown focused on hardware, and as a result, it's not putting much effort to fixing its software. He writes, "Automator, a buggy piece of software with no support, and because it's free, no competitors. Keynote, a presentation program that hasn't been improved in years. iOS 10, which replaces useful with pretty. iTunes, which is now years behind useful tools like Roon. No significant steps forward in word processing, spreadsheets, video editing, file sharing, internet tools, conferencing, etc. Apple contributed mightily to a software revolution a decade ago, but they've stopped. Think about how many leaps forward Slack, Dropbox, Zapier and others have made in popular software over the last few decades. But it requires a significant commitment to keep it moving forward. It means upending the status quo and creating something new." From the article: Software can change faster than hardware, which means that in changing markets, bet on software. It's tempting to treat the user interface as a piece of fashion, some bling, a sort of jewelry. It's not. It's the way your user controls the tool you build. Change it when it stops working, not when you're bored with it. Every time you change the interface, you better have a really good reason.John Gruber disagrees. He writes: Software, in general, is much better than it used to be. Unlike 1995, we don't lose data due to bugs very often. (For me personally, I can't even remember the last time I lost data.) But our hardware is so much better than our software, the contrast is jarring. An iPhone is a nearly perfect object. Sleek, attractive, simple. The hardware is completely knowable -- there are only five buttons, each of them easily understood. iOS, however, is effectively infinite. The deeper our software gets, the less we know and understand it. It's unsettling.
Dude does a bong hit... (Score:5, Funny)
...ruminates on how it's not the bong that matters, but the weed that goes into it.
Stays up late writing blog post on same idea, but extrapolated to hardware & software.
Re: (Score:2)
what is software without hardware to run it on?
i think all he is just saying our grasp of hardware is way better than software because software has less control.
Re: (Score:2)
Sure, the VCR could implement internet connectivity to retrieve the time from an NTP server, or get the time from the broadcast stations, but no matter how much code you put into your firmware, without that hardware implementation, IT WON'T HAPPEN.
Re: (Score:2, Interesting)
That user interface is written in SOFTWARE.
The software, in this case, is an afterthought because the UI was designed by hardware guys. What mattered to the VCR designers was to keep the number of buttons low and therefore cutting costs.
Re: (Score:2)
Remove a horse's brain, and neither the brain nor the rest of the horse will win many races.
Hardware is so much better? (Score:3)
When I was a kid and turned on a BBC Micro, it was ready to use instantly. Same with the old TV I had. And I could watch anything I wanted to watch on that TV, whether it was from the aerial or the computer or the VCR. And on that VCR, I could just fast forward through any initial stuff on the tape I wanted to skip. And some of those devices worked for a decade.
Today's world of hardware that costs hundreds or thousands but fails within a few years, if it even gets that far, is not an improvement. Today's wo
Re: (Score:2)
> Today's world of hardware that costs hundreds or thousands but fails within a few years, if it even gets that far, is not an improvement.
With the right software hardware from a decade ago is still running. Stop buying the cheapest hardware you can find. Look for embedded systems for industrial environments. They might be a bit slower than what you're used to but they'll run for the next few decades just fine.
Re: (Score:2)
When I was a kid and turned on a BBC Micro, it was ready to use instantly.
And for its time, it was awesome. And today, a pocket calculator makes that BBC Micro its bitch. But the truth is that most of us never turn our computers all the way off, so it doesn't matter much what the power-on-to-usefulness time is unless we're experiencing a lot of crashes.
Re: (Score:2)
I don't typically turn my TV or PVR all the way off during the day either, but they still take forever to start showing me programmes.
Re: (Score:2)
Sounds less like a problem with technology and more a problem with cultural attitudes of those who had it engineered. The 80s/90s had the last bits of pro-empowerment (real empowerment, not socjus 'empowerment' as it's typically defined today). The whole idea that computers are there to be useful, powerful tools controlled by the user got turned on its head. Now it's all about corralling the user into online 'portals' and charging monthly fees. This way, 'unauthorized' actions can't be taken on 'unauthori
Re: (Score:3)
I think you're suffering from a bit of rose-colored nostalgia.
I remember cars not starting on winter mornings because they were temperamental as hell, and breaking down much more often, requiring costly servicing or repairs. By contrast, today's cars run far more reliably than they used to. I've heard people complain about all the electronics packed into them, but it's all those electronics, among other factors, that keeps the car running in good condition and warns you when anything goes wrong. Many mod
Re: (Score:3)
I'm afraid we might just have to agree to disagree on some of this.
Certainly you're right that modern cars are more reliable, and the better built-in diagnostics are a part of that. But the flip side is that you used to be able to buy a repair manual for any major model of car and take care of it yourself, and if you did then many popular models could last almost forever. Today it's barely possible to change a light bulb or diagnose the cause of a simple warning light in many new models without going to see
Re: (Score:2)
What you are describing is called "forced obsolescence" and it's nothing new, hell it predates (personal) computers. G.M. invented it as a businessmodel back in the 1960s. The computer world copied it by the early 1990s. Remember when the i486 was the flagship intel CPU ... for over a decade ?!
From the Pentium onwards they were bringing out new models almost every year with the previous year's model basically unusable within 3 years (partly because software requirements would chase the latest and greatest).
Re: (Score:2)
Yes to most of what you said.
The only thing is that the new stuff is way, way cheaper. If you buy a premium brand now, the cost will maybe be the same as a normal brand in previous decades. Back then the cheap low end didn't exist.
Re: (Score:2)
I'm sure you're right and that's part of it. My worry is that we'll go so far that you can no longer buy decent quality products at any price, even if you want to.
Re: (Score:2)
Re: (Score:2)
Indeed. Don't forget how important it is to seal up things like phones as well, so it's extra-difficult to extract and recycle any rare elements inside.
Re: (Score:2)
Well, I still have to disagree, at least to some extent. And not with a -1 mod (I hate that too, btw).
I think the biggest difference is in whether you buy bottom of the barrel priced and quality stuff or not, even with computers. For example:
I purchase my computers from a custom PC boutique dealer and probably pay half again as much as a comparable brand from a box store, maybe even more. But these guys analyze each component for failure rates out in the field, and only sell the highest-rated parts in te
Re: (Score:2)
I wish my experience were similar, because I'm also the kind of person who doesn't buy cheap tat and does do his research. I only buy from reputable sources. I typically buy mid-range products at minimum, and often towards the higher end. And I have still encountered dramatically more failures generally but also dramatically more deliberate crippling of products in recent years.
I do agree that there is some element of modern technology simply being more complex and/or working on smaller scales and so inhere
Meh (Score:1)
An incompetent VC idiot who lost a quarter of fund's portfolio value in five years tells everybody how to make money... hahaha
Re: (Score:2)
An incompetent VC idiot who lost a quarter of fund's portfolio value in five years tells everybody how to make money... hahaha
VC was all about hardware lately. Fitbits, VR, IoT/DDOS thermostats. Money was not made and startups went down in flames, so now we have experts saying it's all about software.
It's just like in finance. Stocks, bonds, PE. Stocks, bonds, PE. According to the wheel of fortune we're heading in stocks territory.
Re: (Score:2)
I just love how the examples of amazing software innovation given are Slack and Dropbox - Web 2.0 versions of IRC and anonymous FTP.
How about an original idea? Do those even exist anymore?
O RLY? (Score:5, Insightful)
The hardware is completely knowable -- there are only five buttons, each of them easily understood.
If he knew about the hardware, he would know the action of every button is software defined!
Neither of these fools understand hardware or software beyond a superficial measure.
Re: (Score:3)
... and if he knew anything about the hardware he would also know it gets just us subtle, and in fact a lot of hardware has microcode, which is of course software, but encapsulated in a hardware interface.
Re: (Score:2)
... and if he knew anything about the hardware he would also know it gets just us subtle, and in fact a lot of hardware has microcode, which is of course software, but encapsulated in a hardware interface.
I think you mean firmware. microcode is different because it actually changes the behavior of the IC in a similar fashion to an FPGA. microcode is used by very few ICs.
Re: O RLY? (Score:2)
Money drives innovation (Score:5, Insightful)
Re: (Score:1)
InB4 the typical /. response that "software replication cost = $0. Therefore, software should be FREE!" Never mind that hardware replication cost in not very high, either. The R&D cost is very high in both s/w and h/w.
Re:Money drives innovation (Score:5, Insightful)
If we want innovation in software, we have to be willing to pay something for it.
I couldn't give a toss for innovation. They can blow their innovation right out their arseholes. What I want is iteration. I want them to go back over their work and fix their mistakes. I am willing to pay far more for a bugfixed OS than I am for a new scheduling API for example. And I won't pay anything for the developer to add spyware to the system.
Re: (Score:2)
People will fork out a few hundred every few years because the last trinket they bought has been obsoleted by the manufacturer prematurely, either by making the latest software run like shit (Apple, Google Nexus) or by never releasing updates to begin with even if there are egregious security problems (most of the rest of Android).
Hardware can't be remotely upgraded over Wi-Fi - there is a measure of lock-in there that cannot be dealt with.
Re: (Score:1)
If it were free software, you or the free software community actually had the option to improve it.
Whether it be free software or not, the parent is of course right that one should not expect a lot of improvements if not paying for it (however, see reply #53142799).
Re: (Score:2)
Let's see...users are willing to fork out a few hundred dollars every few years for the latest tech trinket, yet they want all their software for free.
I'd like all my hardware for free too. If hardware was free to make copies of, I'm sure it would be in the same boat. This is what economists call being a rational actor [wikipedia.org]. If I don't want to fork out $10 for something, then clearly having it is not worth $10 to me. This is just basic economics. Arguing against it is no more sensible than raging against the incoming tide.
No both matter (Score:2)
Hardware may be a "solved problem" for x86 cpus, but anything beyond that is still unexplored territory. Elon musk is sad that so much talent is focussed on some cloud solution while that talent is terribly missed engineering. I agree with that. We need electric cars that can replace ICE ones, we need robots that can take care of the elderly, and this is only partially a software problem.
Re: (Score:2)
Bottom line: The elderly hate everything.
Re: (Score:2)
The elderly don't want robots to take care of them.
The elderly don't want robots to take care of them, but the definitely won't want to pay more for a real person to take care of them when robot care is cheaper, so robots will do -- but they reserve the right to complain all the same.
Re: (Score:2)
Software is the story.
Shingy, is that you?
Why write software? (Score:2)
When you don't have to. Stick to your core competency. Software is notoriously hard to manage precisely because it is nebulous. You cannot see it, touch it, feel it, smell it, or taste it. A strictly intellectual construct. Hardware is easier to conceptualize as it is tangible. Does apple really need to redo MS Word? Or Docker? Or a host of others.
Sure software can change faster, but all that does is create crappy software. Who hasn't experienced sales screaming "We have to ship feature X before our compet
Re: (Score:2)
Not really. If you truly understand hardware you know it is a boundless discipline, and in fact there is often microcode embedded in said hardware. It's all about the interfaces and their integrity. Software, hardware, it doesn't matter.
Staff have to be smart again (Score:2)
Video is done at HD or 4K with free software or low cost solutions or software from a big hardware brand. Good enough to upload and share.
Music is streamed and created by creative people or a group created by committee to sell well.
VR is still been hyped as not inducing user issues and needs a gpu and cpu to get the frame rate.
The problem now is branding and the optics of the b
Re: (Score:2)
Why was in house dx12 support not ready?
Ugh. DirectX is only even a thing because of 3dfx. OpenGL? Never heard of it! Here's this GLIDE thing. Wait, Microsoft is going to make their own 3D API? Shit! Here's a limited, half-assed OpenGL driver! tooooooo late.
Re: (Score:2)
More brands have to look at who they are adding to they own staff and why the products fail.
Re: (Score:2)
Re: (Score:2)
DirectX became a "thing" because of "The Lion King" on PC.
Whoops! Typo. I meant Direct3D. My bad.
Seth Godin is spot on with this one. (Score:5, Interesting)
He is, IMHO, 100% correct with his analysis, including the critisism of the quality of what made Apple great. Apple abandoned their opinion leaders (us) about the time they started requireing a sign-up to get the devtools. Slowly but surely their Unix isn't quite that attractive as it used to be and the quality of their utility software has been in steady decline ever since. The last few versions of Preview can't even render PDFs correctly anymore.
Meanwhile the open web, pushed by Google, is taking over. Devices and web environments are steeply growing in power, and the line between website, service, VMed and native app is blurring faster than we can follow.
I've been seeing it ever since I finally understood ChromeOS.
Remember when it came out? Everyone, including me, was like "WTF?".
But now we understand. Chromebooks are the poor mans and the developing worlds (80% of all potential users globally) MacBook Air. They're dirt cheap, boot nigh instantly and run for a day on one charge. And Google takes care of you all along the way.
Today it's blatantly obvious that Google, of all megacorps, has the best long-term strategy and thus is pushing a standards based open web. It's the only plattform they can win with and it is more and more becoming the plattform with which people can develop safely and be guraranteed some sort of userbase, no matter the underlying OS or device. The Pixel comes as a premium phone - an unusual thing from Google - but everyone knows it's just an upgraded iPhone knock-off hardware wise. The real deal is with Google Assistant and the unlimited storage they offer.
As for the web being the plattform that is evolving the fastest - yes, of course it is. Updates are as close as refreshing a pageview and storage and AI are dropping in prices and power in huge leaps as we speak. I've been torn to and fro about wether I should leave the web for some 'real' programming and environment ever since I switched my career into it 16 years ago, but I have to say that it never has been as interesting as it is now to stick with it, sit back, and quietly watch as the toy language JavaScript takes over fields no one ever even dreamt of 10 years ago.
My 2 cents.
Re: (Score:2)
Seth's glorified blog post is not news - and it is barely noteworthy. He is looking to Apple for innovations in word processing and spreadsheets, as though that were the new frontier? That is laughably out of touch.
He claims no progress in "file sharing and internet tools" ... right after Apple simultaneously releases mobile and desktop operating systems that automatically share a common clipboard, documents folder, and desktop across all devices. Whut?
And this is on top of the "file sharing and internet
Few decades? (Score:2)
"Think about how many leaps forward Slack, Dropbox, Zapier and others have made in popular software over the last few decades"
None of these are even a decade old! They are cloudy web x.0 incarnations of old unix utilities!
And Seth should know this! He is as old as I am!
Now get the hell off my lawn!!
Re: (Score:2)
Yeah, I'm not sure how he gets "omg software innovation is amazing" from Slack or Dropbox.
Look! We can make IRC and FTP pretty! Innovative!
Chess is effectively infinite (Score:1)
No software crashes? (Score:1)
He can't have used web-browsers much.
Since Chrome goobles up whatever amount of RAM eventually it will be too much and you will want to kill that fucker.
Twice in a short time period Vivaldi (Chrome based by guys from Opera) and Chrome made Windows either by itself or through swap ~1TB of data each time to my SSD.
No, its about the Trinity ... (Score:2)
This is the holy trinity of computers:
* Hardware + Software + User Experience
Great Hardware enables great software.
Great Software empowers the user experience.
Great user experiences has people loving computers.
Crappy hardware can only make for poor software.
Poor software makes for laggy user experience.
Laggy user experience has people hating computers.
--
If you're frame rate (or UI) is not targeting at least 60+ Hz, you're doing it wrong. Only amateurs target 30 Hz.
Re: (Score:2)
Thanks for the catch -- yes, should be "your" -- I had a brain fart when I changed the grammar.
Apple as example: Not by accident ... (Score:2)
They learned that lesson because they sometimes did step in and kill interest in developing for the Mac (more so with OS9, but the lesson stuck).
i
Seriousl? (Score:2)
"An iPhone is a nearly perfect object. Sleek, attractive, simple."
*Gag* Really? A "perfect object"? This ain't no dodecahedron we're talking about here. It reminds me of an old soap dish. Stop with the worship of cheaply designed disposable tech. It's a bad mindset and bad for the environment.
Sleek? so thin the damn things bend/break all the time. Oh wow, such engineering. How about we go after "battery lasts a week between charges" rather than "Can be used to jimmy a door"
Attractive? I don't think so, pers
The hardware is completely knowable? (Score:2)
Umm, sure 5 buttons, one of them being a fingerprint reader. Oh, and 3 axis accelerometer, multiple thermometers, a magnetometer, a microphone, a multitouch touch screen, a couple software defined radios, a lightning port that does a more than just USB, whatever else I forgot. All capable of being inputs which can control things in the phone. I think maybe Mr. Gruber was fooled by the sleek exterior and thought he knew the hardware. The hardware is so unknowable that there are forum discussions about othe