Slashdot Mirror
Woz Says Big Software Doesn't Work
chrizbot writes "A friend of mine studying journalism at Google's alma mater interviewed Steve Wozniak of Apple Computer fame. He chimes in on open source, DRM, record companies and how software from big companies suck so bad (including Apple's!). The part my friend doesn't include is how he guessed a trick was performed and won a necklace from him!" From the article: "Sometimes the engineers are true artists and really care what they're doing, doing a really great job. Although, I don't know how much I can even say that because the big companies, Microsoft, Apple and AOL, they tend to turn out the crappiest products, you know, software-wise. The ones that have the most bugs, the most items that are supposedly in there but don't work. The most things that are left out because they aren't finished. The most things that are inconsistent with the way they did their last program. I get the worst, worst software almost always from Apple."
"Sometimes the engineers are true artists and really care what they're doing, doing a really great job. Although, I don't know how much I can even say that because the big companies, Microsoft, Apple and AOL, they tend to turn out the crappiest products, you know, software-wise. The ones that have the most bugs, the most items that are supposedly in there but don't work. "
It's a symptom of two things, from the standpoint of poor quality software produced by people who are capable of much better:
1) Nothing personal at stake for the people actually producing the software. It's a lot different when your livelihood directly and visibly depends on the quality of the product your employer produces. Whether it's because it's my own company, or I get fat stock options, I'll work harder when I'm trying to reach the cheese.
2) Diluted responsibility for the product. 2,000 people working on a product means that in all likelihood, my individual contreibution will go unnoticed, and therefore I have less incentive to perform well. Also, even if my contribution is perfect, it won't have that much effect on a huge project.
"Trolls they were, but filled with the evil will of their master: a fell race..." -- J.R.R. Tolkien on Olog-hai
My view has always been: don't let developers (including me) use the latest & greatest technology. Force the build once a week to be run on an "old" PIII @ 800 Mhz w. 128MB RAM. If it's un-usable for quick testing, then go back and fix it.
(by the way, I know I'm being generous in those specs, I personally test all my software on a dog-slow Pentium II @ 233Mhz w. 64M RAM running various "older" OS versions (Win2000, Linux 2.2!, etc.)
Then, when you roll it out to your users and their running the latest 3GHz, 4GB RAM machine, they are happy.
Linux & GNU seem to be the latest (last five+ years) culprits in the bloatware regime. I remember actually compiling the full kernel on an 8MB machine (yes, it took four hours)...now you can't do in under 32MB
(although I guess that's more GCC bloat than anything)
Things are just too big and bloated now.
Give me an old "Classic" Unix with no X, just command line.
Let me pipe my various home-built tools together to create a final simple working FAST result.
TDz.
break with the "tradition" of their insolance and:
... just the agregate personal fortunes of the top swindlers who perpetrated this biggest scam in the history of the world.)
1) offer customers a sincere apology for their negligence
(no court seems able to get a comprehensive conviction
against any of them anyway, so they should't have to worry
about liability), at the same time as
2) distribute a genuinely effective set of patches to those
customers as they wait for the company to develop a new
product that actually does what it says it should
3) distribute that genuinely secure product to customers FOR
FREE, with full on-site support to smooth the transition
4) offer a discount on upcoming products to extend good will,
5) and eat the crow they so richly deserve.
I figure all in all it will only cost them about a trillion bucks. (Yeah
But why should the software industry be held to a different standard? Other industries end up eating losses all the time.
Because if all those middle/upper-level managers in other sectors that wanted to cash in on the "replace workers with machines" craze of the 90's would propose such an ultimatum to their companies' stockholders, then they would have to admit that their blind greed backfired all over their damned faces!
Points I came away with:
-Apple is no longer what they were when they started out, and now their proprietary software sucks, the 3rd party Apple MAC software is great, and the Apple Macintosh software is great
-Proprietary software traps you
-Open source is good for companies that would like it, but Apple software is still better
-DRM is a necessary evil in the digital downloading world, since people share files and hurt the artists
-CDs and Itunes should be cheaper, artists should be able to set their own price
-Software is huge, complex, over-hyped and under-supported and it is only going to get worse
-Colleges should train people to design software with a humanist point of view
After reading this article, you could argue that the computer industry is quite depressing if you start to think about all the different things he has mentioned. If you want to build a better computer/OS/hardware/software, you should not put large corporations in charge of development, leave it up to those with a more humanist point of view. The only problem is, if by humanist you are saying it is for the greater good or some moral good, it is inherently against the profit model and the actions of greedy corporations who are always trying to increase profits or meet projected profit expectations and deadlines.
The Open Source community is the closest thing you can get to a 'humanist' point of view while computing. Since the profit motivation is taken out of the equation, everyone can benefit.
He who knows best knows how little he knows. - Thomas Jefferson
Well, here's some balance for you : The Guardian on why OpenOffice sucks so badly.
Athletic Scholarships to universities make as much sense as academic scholarships to sports teams.
Before you paint *all* open source software with such a broad brush, realize that the fink repositories are open source and that OS X userland programs are based on open source. In fact this message you're reading has been served up on an open source webserver and has probably passed through several routers running open source software.
QT 7 on Windows: There is no way to make it ignore the proxy you set for Windows. QT 6.x has a "Streaming Proxy" panel in which you could just uncheck all the boxes and be done with it.
Not hardly. I've found OSS software has plenty of things/features that don't work, or don't work they way you'd think. Often, its because some package is still in early stages of development. People often install a linux distro with the impression that the *entire* distro is a finshed product, which isn't the case. Installing a linux distro is a different situation with respect to where various parts of the distro are at, and can be frustrating due the amount of information that needs to be assimilated to get a perspective that helps dispell the confusion.
That said, I installed Ubuntu 5.10 on a thinkpad A22m, and I've only had one thing fail to work, minicom which doesn't talk to the serial port, and epiphany crashes from time to time (although it works). A quick laundry list of things that pretty much worked fresh out the install without a hitch:
The following I just built from source, in the most thoughtless
I installed OSX 10.4 on an 800MHZ iLamp, and it crashes, and the mouse occasionally stops talking with the USB port - none of which ever happened on 10.3 - so its the software. Apple QA does seem to have taken a hit lately.
OSS 1
OSX 0
I have to say, I think WOZ is right.
"We are all geniuses when we dream"
- E.M. Cioran
Woz Still Loyal Apple Zealot
From the article: "...I love every part of the Apple world. The whole world of Apple works together."
Trust me. This is an inactive account. Regardless of what the
Thank you. Someone actually brought up user interface design. I think that is what woz was getting at. He was talking about the raskin's of the world that innovatived the desktop. HCI (human computer interaction) is a field of study that most people don't think about anymore. I recently took a class on gui design at my university. Prior to taking the course, i actually had a cs professor (my advisor) recommend I avoid the class. He didn't see a point to the class. I certainly did. I don't plan on devoting my life to HCI like my wife does, but I certainly think its beneficial to web designers and conventional application developers. Usability is quite important. To this day, whenever i write an app or design a website I test it with the mom test. If my mom can use it efficiently and understands what each feature/command does, I did well.
I like the OSX dock provided magnify is disabled. Resizing the icons makes it harder to move your mouse to the correct one. Its annoying.
I don't think I would agree that OSX is the worst operating system in terms of usability. Solaris comes to mind. CDE and the Java Desktop System are crap. There are so many things you can not do in the gui that require CLI interaction its not funny. I love CLI interfaces, but I can't ask my mom or even my boss (a novell guy) to use a unix terminal. They freak out. My favorite OS of all time is NEXTSTEP, but I wouldn't recommend it to everyone. It seemed so consistant compared to its modern counterpart (OSX).
Skinning apps is stupid. Not only does it lead to inconsistancy, but it also eats up memory like crazy. People who love it are the same people that complain about their computers being slow or talk about buying an extreme edition p4 just to run winamp, IE, and a few games.
I think most people are bad at UI design. Very few people at my university seem to understand basics and even worse most don't even think about usability when they write software. I know I'm a bad offender, but I try to improve. I think thats all we can ask from Apple, Microsoft, and the rest.
As a roadmap for OSX, I recommend the following:
1. Fix kernel/stability problems.
2. Consider standardizing OS components on one UI or at least limit it to two. Pinstripes, shinny metal, and now the worst of all.. plastic. Why do I want to look at plastic? Mail.app drives me nuts.
3. Ignore adding 200 features and work on getting the system consistant, fast and usable. Thats a feature in itself. With the intel switch coming up, I know we are going to have major stability problems in intel and ppc based macs.
MidnightBSD: The BSD for Everyone
Please note that I did not call iTunes the best music playing app out there.
I called it "the only one worth getting excited over." From the heat of the debate which followed (to the extent that everything else I said was completely ignored), I would say my statement is pretty well supported. Who would ever spend this much time arguing over the UI decisions made by Media Monkey?
Lack of WMA & OGG support, and an "ugly" interface are both valid criticisms, if such things are important to you, but iTunes stands out for several reasons:
1) AAC. It's a standard codec, superior to MP3 in several ways, which many other players are lagging behind in supporting, and the ideal codec for use with an iPod (which is the same as saying "the ideal codec for 90% of the MP3 Players out there.")
2) Apple Lossless. Yes, there are a couple of other players out there which support FLAC, which is another lossless compression codec. If I own an iPod (see point one), guess which one I'll want to use for music I don't want in a lossy format?
3) iTMS. This is huge. An "a la carte" music store seamlessly within the application. Since it came out, I flat-out stopped looking around at what other music apps are out there.
4) Database-driven file management, with user playlists and auto-synch with MP3 players. Far from the only player which does this, but it does it really, really well.
5) The visualizer. Sure, almost every player has one of these things... and on most other players, they really, really suck. Microsoft's is slightly less interesting than a screen-saver app. When I'm playing music in my media room and turn the iPod visualizer on across my 119" widescreen, people gape in wonder at it. Throw on "Dark Side of the Moon", and it's even better than those "Laser Floyd" shows that used to be popular at planetariums. Seriously, if you haven't spend a few minutes watching it in full-screen mode, you probably are not aware of exactly how well it was done. (And it keeps getting better with each iTunes update!)
As for the "large memory footprint" complaint... What is this, 1991? Who gives a crap?
Information wants to be anthropomorphized.
My guess is that part of it is thanks to people like me, who on the whole like change. The subset of customers I represent love to buy a new release of the software, because we're getting new and more interesting goodies. For us, it just wouldn't seem right if there wasn't some radical change thrown in with the mix, as long as it doesn't slow us down much.
In other words, I think what's going on is that Steve responds to the desires of the user, particularly the user who wants to upgrade and give him money.
But even I will admit, getting rid of key caps was just plain dumb. Nobody's perfect, not even Steve.
But at least he does care, and that's why I stick with him.
D
I was wondering the same thing...Woz makes some pretty heavy allegations, but refuses to back them up with anything substantial. What software is he talking about? I'm a filmmaker and a graphic designer of 15 years, and frankly Apple publishes some of the best software out there...Final Cut Pro, Shake, Keynote, Safari. Granted, all of those apps needed work when they were released (except Shake, which Apple purchased), but they have all matured nicely. That, to me, seems to be the more typical pattern of development for Apple. And who are these supposedly stellar, non-buggy, homerun-every-time, small developers with whom Woz is apparently enamored? Woz refuses to give examples, instead choosing to hide behind the obtuse comment, "it happens all of the time." Frankly, there's a huge difference between developing an app that lets you change your desktop picture, and one that edits film. And as much as I love Camino and Firefox, Safari still renders faster and better. I'll tell you what...if Woz can set up an editing station for me that runs even half as well as my current Final Cut Pro station, using only his brilliant little developers, and gives me even half of the capabilities I currently have, then I will happily jump on his "small guy" band wagon. Otherwise, Woz needs to shut his fat yap.
Umm, thanks for the "mod this up," but I don't think Microsoft does any better. I'm just saying Apple used to be good about this (say, from System 1 until System 7), and now it's bad. But when I went from Windows 3.1 to NT to 2000 to XP, I'd hardly call that a consistent user interface; it struck me that all sorts of things had been moved around randomly. Their system preferences are awful! It's True that Apple rearranged their system prefs over and over again in early OSX releases, so again, I'm not saying Apple's great here, but MS is also really screwed up when it comes to pointlessly rearranging preferences, not to mention they have an absurdly huge and unnecessarily complex set of preferences in the first place. And betraying user interface guidelines? Does MS even have user interface guidelines? I don't know where to look for many common widgets in programs, I seem to have to memorize them again for every program. They're also terrible about hidden features on right-click pop-up menus and such. Anyway, my Mac-bashing doesn't imply MS endorsement. I don't think anyone's concentrating on ease of use in human interface design in a computer operating system.
Can anyone tell me how to set my sig on Slashdot?
One perfect user interface paradigm is not achieveable(sic) because people are all different.
I see better user interfaces in pc games than I do in other software. This is primarily because game designers realized years ago that everyone has their own prefered way of interacting with the input interfaces. This led to the ability for the end user to modify the input options (such as a keyboard mapper and joystick macros).
A better approach for all software would be to build an infinitely modifiable interface - both for visualization and input. Tie the configurations to some standard, like XML. The company would have a 'standard' config file, and users and others would be able to create and share their own interpretations of what is 'best'. Everyone is (or I should say 'has the option of being') happy. If you are not an XML hack - a GUI could be easily built to make the modifications easier.
Most applications do not approach this level of configurability.
Lodragan Draoidh
The more you explain it, the more I don't understand it. - Mark Twain
That's extremely insightful, you don't work with the man, do you ?
My recollection of briefly working for Steve involves a meeting with a group of 8 engineers where he pretty well had everyone convinced that they could, in a few months' time, fully test an entire OS and extensive application suite, on new hardware, while writing a couple of never-before-imagined applications. In short, we were all going to pull off some miracles, pretty much because of a Steve pep talk. It's great to have inspired engineers, and sometimes people can pull of miracles, but that's a scary way to develop products on a schedule and a budget.
My biggest beef with OS X software ( aside from the Finder, which just needs a *complete* re-write ) is the recent lack of UI consistency. Try this : launch Safari, Mail, and iTunes ( most recent versions, in OS X 10.4 ). Check out the look of the windows... are any of them the same? Not really, they're all slightly different-looking... and iTunes looks like no other OS X app ever !
The difference between brushed metal and standard windows was annoying and unnecessary enough, but what is the rationalization for those three Apple-authored applications having such different looks ? Who needs 4 different styles of window dressing on a single machine? They're making Windows look like the platform with UI consistency, WTF is going on at Apple with these differing looks for different apps ?
After reading this whole thread, I've come to a startling conclusion:
You can't please everyone.
These are my friends, See how they glisten. See this one shine, how he smiles in the light.
The best software (IMO) comes from small groups or individuals with exceptional talent, never from a gigantic corporation. The problem with a large corporation is that quality tends to dilute as mediocre people are hired, rot sets in, projects atrophy, clueless managers cut funding, stupid ideas are pushed, brilliant ideas are ignored, problems are neglected and faults are left unfixed for years.
I can reel off dozens of examples. People seem to forget that back when Microsoft was small they made some absolutely spectacular software. The Altair BASIC written by Gates and Allen (more credit goes to Allen IMO) was astoundingly good. Even when Microsoft grew to a dozen people they still produced some of the best BASIC interpreters around. But now that Microsoft is a multi-billion dollar company with 1000s of employees they seem to produce nothing but embarrassing crap.
Same for Solaris; it was way better back when there were just four guys and one of them was Bill Joy. Same for the Mac; it was way better back when there were just four technical guys and Steve Jobs was an over-bearing perfectionist. It's the same reason why Linux (the kernel) still rocks but Debian (the distro) is starting to bite (it doesn't suck yet but IMO the writing is on the wall). Small groups of extremely talented individiuals lead to exceptional software. The larger the team the more likelihood of there being medicore contributors and the overall quality goes DOWN.
One of the things I find most interesting about open-source software is not the cost and not even the licensing, but the promise it holds of building large software projects of higher quality. Companies have tried all sorts of engineering techniques though honestly they seem to get it wrong more often than they get it right. OSS uses a technique more akin to natural selection; the quality varies wildly and some of the mutations are totally crap but overall the quality is continually improving. Whether it will ever produce software that in all respects is better than the "genetically engineered" software from Microsoft, Apple and Sun, I don't know, but I'm keen to find out.
I've been grasping for a *reason* for the difference between these applications... that's actually the main thing that bugs me. I'm not sure I so much mind applications having different basic looks so much as long as there is a *reason*. I've never heard a reason for these differences that made sense to me.
Safari is just plain ol' Brushed Metal, it's different from regular-look-and-feel Cocoa apps because um, it's supposed to be used only for apps that mimic a 'physical device', like I guess an eBook reader or IM-capable phone or MP3 player ? Enough text has been wasted wondering if that makes sense, I won't repeat it here.
Mail looks different because it has this new thing I think they're calling a Unified Toolbar, where a tool bar and title bar are combined ( why? what's the benefit, I can drag the window by grabbing between toolbar buttons, what's that? ).
Now, I don't personally like either of those things, and I find the justifications for them pretty thin, but... what about iTunes? Why does it look different? It has pointier corners and a different skin/gradient, why exactly? The only reason I could think of before now was that it's sharing code with the PC version and as such isn't *really* a Cocoa app in the same way as the other apps, and has started to get a slightly different look... but that's just a theory, I have no idea. Maybe it's prepping for a look-and-feel we'll see with the Intel switch, or maybe Apple just wants it's most important application ever to Look Different(TM).
So your theory about these being different for distinction in Expose makes a *little* sense, and I thought maybe you were onto something for a minute... except that they're all pretty minor differences, really, that are likely to be lost when the window is so much smaller. I'm looking more at the contents of the window in Expose, not the border and toolbar so much. So while it's an interesting theory, I'm guessing that the difference has less to do with Expose and more to do with UI Guidelines being viewed more as flexible guidelines, not as very important rules to always follow... and the differences are due to an 'evolution' in programming tools and freedom given to individual project divisions within Apple, more than anything else. While UI Guidelines are still important, maybe now the religion is about progress and features, not UI guidelines.
The differences are still annoying, at least to a seasoned Apple user and programmer. You just look at it, and you think, "what is going on HERE?" Pick something and stick with it! Geesh!
When is the DoJ going to crack down on Apple for bundling apps that compete with and ultimately destroy 3rd party apps such as Watson and the Caffeine Suite.
It's worth noting in passing that Caffeine Software was basically one guy, who shut it down after he got a job at... Apple. Personally, I have trouble believing that the disappearance of TIFFany at Apple's behest and the appearance a couple years later of Aperture are completely unrelated, but I've never seen any evidence to confirm my theory.
Sherlock/Watson I think they did a pretty dirty thing with; I half wonder if it wasn't that Steve Jobs or someone else high up just got a bug up their ass about Watson's name, which could be taken as a slap in the face for the previous (useless) releases of Sherlock. Konfabulator versus Dashboard, though, I can't get too worked up about; to me, Dashboard is spiritually the return of desk accessories, and implementation-wise, Konfabulator done better.
This is always an interesting dilemma. When Microsoft Word and WordPerfect integrated spelling and grammar-checking into their applications, they knew there was a thriving market for "add-on" programs with those functions, and they had to be aware that their integration would pretty much snuff that market out of existence. Is that sufficient reason not to do that integration, though?