Home
About
Apple
Career
Experiences
General
Graphics
Hardware
History
Humor
Interface
Networking
OS
Opinion
Politics
Programming
Quotes
Reviews
Security
Software
Sound
Thought
Web
Cheap International
Airfare Online
Wachovia online banking
Get Free Coupons Online
Finding the perfect
discount hot tub
Payday Loans
Stock Trading Online
Stuffed Animals
Smart Investing Online
|
|
OS X Jaguar : Whats cool?
All the things I like about OS X
By: David K. Every
|
Kind:
Created:
Size: |
Article
2002-09-13 16:00:01
21 KB |
|
|
 |
 ith Jaguar released, I was pretty happy. There are a lot of improvements in it, and progress is obviously being made. Then I hear the people saying that “it is a good as a Mac”, or that Apple is killing MacOS 9, so I wrote a whole article listing all the things still missing in Jaguar (here). This isn’t to bash, but just correct those who mistakenly believe we are ahead of the game. We are overall, but not when it comes to Mac OS power users (and interface), yet.
The problem is that while all those things missing are valid complaints, without the balance of all the things that I like in OS X Jaguar, you could get the wrong impression and that I can’t stand the thing. I like it and I use it as my primary OS, and have no intentions of going back to Mac OS. So while Jaguar has some holes and shortcomings, that I listed, there’s a long list of things that I do like about it. And I wanted to list all those too, so that people can see the many positives, and to balance out the other article. Here’s the good side.
   
OS X (Jaguar) is cool because UNIX is cool. Not because UNIX is a particularly well done OS, UNIX has tons of anachronistic design choices and has plenty of legacy issues that aren’t pretty or modern. I’ll go into UNIX issues more in another article. But UNIX does have many strengths to counteract those issues.
The biggest strength of UNIX the size of market, and the size of market issue is much more significant than pure numbers; the UNIX types tended to be around a few markets, like development, OS development, and network administration. So they are not only large markets, but also significant markets, and new markets. UNIX is test-bed for most of these technologies. If Apple wants to move a new OS forward, they are going to be borrowing from UNIX anyway, so why not make their OS UNIX derived?
There is also openness about code (open source), and API’s and design that has permeated UNIX. Many have contributed to UNIX, and still do. When you give it away for free, others use it. This has snowballed for many decades. UNIX evolves, and you can stay closer to the cutting edge if you don't have to port everything to a different platform.
UNIX is also very stable, in almost all senses of the word. It can be a bitch to setup, but Apple is hiding much of that. And once setup and configured properly, it will work, and can survive bad software practices (like poor QA) pretty well. You can set things up and just leave them, and know that when you come back, they’ll probably still be running. Better than Windows, far better than Mac, and nearly as good as IBM mainframe type solutions. This stuff is robust. On a desktop and in many other markets, this is going to make a significant difference.
However, when I mentioned stability I also meant it as in “not changing”. While little things evolve in UNIX, and change constantly, it has been UNIX for 30+ years (arguably 40). In many ways it takes a lot less time to change from versions of UNIX than it does between versions of Windows. That stability is incredibly comforting to many programmers, network administrators, academics, researchers and just plain users. They want to know their machines; and sometimes they have had a longer relationship with them than anything or anyone else. Learning a new UNIX takes them a few hours, days or weeks, depending on level of intimacy they want, but they then know all sorts of things in incredible detail. UNIX plays to human nature and the dislike of change, and the thirst to know, and the drive to have control over one's destiny and environment. UNIX does that better than any other OS out there. OS X can ride on that. And users know that whether OS X lives or dies, most of the knowledge they gain learning about UNIX can be applied in the future. That’s emotional stability for people, and they love the platform that provides that.
         
Now in the previous section, I was talking about code as far as the OS itself. You could get lots of software from others, it was open, it did lots of things, and they had good people. But all that stuff also applies to Applications and Application developers as well. UNIX types tend to be clustered around a few markets, development, academia (higher education), research, network administrators, and some vertical markets in the high-end arenas (high end video productions, high end animation and 3D, and so on). But these are all people and markets that Apple wants. So again, who (or what they do) is as important as how many.
When it comes to writing applications with UNIX, you not only get whole apps (commercial apps that are easier to port to Mac because most of them is already written), but you get new Apps and parts of Apps. Many people write apps, because they borrow parts of other apps to build on first, and put their changes out there for others, and so on. So momentum builds momentum, and more Apps mean more customers, and customers are good.
This openness makes it cheaper and easier to write for UNIX than competing solutions. While a lot of the free code is crap, a lot of commercial code is crap too (and usually far worse). At least in the open markets there is slightly less pressure to ship it yesterday; so often they will take more time and do some cleanup and restructuring now and again. And having lots of eyes on code just means that more programmers can fix things, and knowing that other are going to see your code keeps some programmers more diligent. It is hard to say what is worse or better, there are stellar examples of "how not to" in both. But if you're starting something new, having a head start, even a bad one, is often better than nothing.
Again, UNIX is cool, not because it is particularly good, but because there are a lot of UNIX developers that only know or will only consider UNIX. They bring Apps and give away code. They are porting their Apps to OS X. Some of their Apps are quite good. This openness also means standard tools and standard apps. I’m doing WebObjects, PHP, some perl and cgi scripts. None of this was done, or done as well, on Mac OS 9. So while technically possible before, most weren’t expending the effort to do so, and it was unlikely to happen without UNIX. I know if I bet on UNIX, and these UNIX technologies (especially the open ones), that it will be there tomorrow, and forever.
For those Be advocates out there, this is something that Be wouldn't have brought to the party. Be had some advantages: but NeXT brought UNIX. There are serious tradeoffs with doing that, but I think it was the right move.
  
People often over-stress the impact of kernels and ignore that a ton of mission critical software (like airplanes and satellites and so on), often do not have a preemptive kernel or full memory protection and so on. Good QA and software practices can compensate, and there are layers and designs that make such things less needed: look at a language like Java that has its own memory management built in, and so on. So there are many ways to crack a nut and solve a problem. All that being said, there’s still a lot that a good kernel and foundation can bring to an OS. Most importantly, since people weren’t solving the problems any other way, and they needed to solve it, then they needed to solve it this way.
Basically a kernel offers low-level memory protection, scheduling and communication between tasks. That’s it. It ain’t rocket science. Having it is usually better than not having it. And bringing it to the Mac has meant more stable apps, and the potential to do more at once. Both have been realized. Jaguar is still slower at a single App than OS 9 for many things, showing that there is an overhead for kernels. But running many apps and services at once, and you gain any losses back. Running multiple processors and you gain even more. And the more complex things get, the better the returns. I don’t know about you, but I’m willing to pay some minor performance overhead for stability.
Modern memory management means I don’t worry about Application sizes. It also means that from a programmer’s point of view it is far easier to write Apps and not as much manual control of memory (less code). In fact, coding is a lot easier without all that nasty manual memory management stuff, and cooperative scheduling. So making programmers more productive means better programs, fewer bugs, and more companies that can and will write apps. Users like that.
The Mac had low-level communications; some better than kernel threads; so not a huge win there. But doing them standard ways means more apps just run (without work), and they run faster.
  
This area is the most mixed of all; with some huge wins and still pretty annoying losses/omissions. I go into the losses in the other article: so I’ll look at the wins in this one.
The new PDF based imaging engine is good. Fantastic quality. I think that we got it as much because of NeXTs NIH (and it was closer to what they were used to), than anything else. But certainly it is more open than GX was. I do think GX had some better typography features, some UI improvements and print features, and the more object based imaging was “interesting” (even if implementation had some issues). But overall, Quartz is way cool, and enables some great features. Some are misused (can you say translucent menu’s), but most are wins.
One win that is just awesome is shadows. I know it is a little thing, but I think casting real shadows on cursors and windows fits the spirit of the interface much better than the little two pixel black thing did; but it just wasn’t possible until Moore's law made it possible. And I think Quartz Extreme, while being an Extreme memory pig (orders of magnitude beyond QDGX) is a good furthering of those graphics, and is a step in the right direction. I would like to see more 2D rendering acceleration through the video card as well: but compositing is a big win, and really improves the smoothness of the interface, and offloads the processor nicely.
So PDF, while being more closed to old Apple technologies, is more open to the market. OpenGL follows that same theme. I’d have liked it if Apple tried to borrow more from QD-3D and put it into OpenGL (and feedback into the community); and not doing so felt like NeXT’s NIH rearing its tyrannical and ugly head. But overall, OpenGL is the right trend, “Plays well with others”, and the implementation has been very good.
The controls vary. I keep thinking about that stupid little combobox, and thinking “nooooo!! Don’t you guys understand Interface at all?” That’s a huge step backwards. But most of the other controls are good. Some additions like the little pimple (blackhead) in the close box, to tell you it is ripe and you need to save before it pops is excellent subtle user feedback. Someone should make an auto-save function called stridex. I always thought it should be a little more informative or look like something that fits a metaphor, (a pen, a disk, something like that and near the close), but it works well either way; people get it, and use it.
Even the general look of Aqua is very nice. It is clean, bright, and usable. I like anti-aliased, even if I war with it (rarely), it really is a better use of a display overall. Aqua is a little wasteful of screen real estate, but not that bad. And as pixel densities and resolutions go up, the room for complaints about that goes down. Sort of Moore’s law applied to imaging and interface. While Apple had some “resolution independence” built into QD (and more in GX), it had issues and was never fully utilized. I believe that Aqua and Quartz gives them far more possibilities. And I'd love to see 150 - 200 DPI displays, with some dynamic reduction capabilities. Quartz also has far better 1:1 mapping between the on-screen and printer imaging: which fully fits everything that is good interface and the goal of the Mac (WYSIWYG).
 
- The Finder is still one area that still feels like it was designed by NeXTies trying to mimic a Mac, rather than by the real Mac people. But they’ve gotten much better, and are slowly learning about details. Still, I rant about all the little details with the fit and finish stuff that I miss in another article. But ranting aside, it is getting better. And I use it a lot, and am noticing the bad stuff less and less, and the good stuff more and more.
- And there are wins with the Finder as well. I think Column-View is far better at navigating deep hierarchies and exploring some areas than some of the Mac ways. It had holes without spring loaded folder behaviors: but that is half fixed. And they’re even getting a lot of the little holes with how or where windows popup (and in what state) worked out.
- The back button and control in the Windows as options are a win. Good stuff, that people use. As much as I hate to say it, I think Microsoft was right in making the interface feel more like a browser. People are becoming used to the browser metaphor, and so borrowing from that makes sense. And Apple borrowing from Microsoft there makes sense as well.
- The Open/Save dialogs has plenty of issues. But I do think that mimicking column view is a move in the right direction. If we can make it consistent, then it will be fine.
- Long file names are a win, as are Unicode ones. Apple was going to have that working either way: but sooner is better, and they work (more or less) now. A little more time to polish the interface (and not throw errors), and it’ll be great. I do think old Apple was smarter with 64 Unicode characters as the limit, and think 256 characters are too much like a book. I would have liked to see a special field to describe what a document was added and for people to search that (sort of an advanced label thing); I think BENTO (used in OpenDoc) had some of that; in fact its journaling, autosave and object stuff was awesome as well. There's a lot we lost there, but things like that might come back. And at least we don’t have a 32/24-character limit any more.
- I like some of extension renaming feedback stuff. It is good that it throws an alert with “you are changing the type of this file?” The extension hiding and using it in place of type/creator is just wrong. But Apple is correct in that they are doing a slightly less lame version of extensions than Microsoft did, and it is the best UNIX UI implementation I've seen.
- I like have a CLI. Sorry, I'm a geek, and having a command line is convenient for some things. It is bad to require it, nice to have it. MPW worked, but was slow and non standard. UNIX gives me many shells, that are fast and standard. If Apple could give me a better interface than terminal, and more interfaces so I could avoid it even more, I'd be ecstatic.
- Speaking of MPW, having better dev tools is a win as well. I have a love/hate relationship with Apple's project builder and other tools (WebObjects). And while I sometimes curse at them and their interface quirks, I do think they are a trend in the right direction overall.
- The menu reorganizing was an improvement; the world: application: file/document domains make sense. Even the little control-strip things stuffed in the menu bar make some sense; they always behaved similar to menu items, so putting them there isn’t wrong. My concern is about space (my menubars are getting crowded). Even the shadow on menus, the font changes and the white space instead of lines are improvements.
- Large icons are nice. I would have probably gone for a vector based solution, and stuck with “animation” style rather than “real life” photos, as the former conveys abstract ideas better. I miss a better rules set on how/why to use icons; we lost some of the lingo-franca of icons. And the big icon-happy interfaces get on my nerves sometimes. I also wish that they were more consistent in showing detail: why do documents show their content but .gif and .jpg’s not? So there needs to be fit and finish work. But the new icons look good, and they do convey more feedback and information -- that's a win.
- I also like that Apple is willing to keep pounding on things. I thought for years about old Apple, “why can’t you keep improving the calculator” and so on. It isn’t like it is that expensive for Apple to have just one guy who occasionally puts some time into the Calculator to add things like scientific functions, tape functions, or HEX support, and so on. This is cheap and easy, and positively effects many users. New Apple is perfectly willing to do that, and that is a critical step in the right direction.
- The iApps are just the right idea. Many don't consider them part of the OS, but I do. Operating Systems are a set of services to make programmers and users lives easier. There have always been more "utilites" creeping into OS's. iApps take that to the next level, and Apple just does that stuff damn well. All of the iApps have holes and need polish, but new Apple is doing them at all is a big win, and I know they will keep pounding on them and making them better. So iApps make an off the shelf computer (and OS) far more valuable to me than anything else has in a long time.
- I like the sherlock stuff (even if it was ripped off, and they screwed a third party developer to get them). Don't get me wrong, Apple was right to add it to the OS, but they should have just offered something or been a little nicer to the guy they screwed over.
- The one thing I miss most from NeXT is having some reference stuff built into the OS. (Can you say dictionary?). I think global dictionary functions are very important services, and would like to see that globalized. I like the sherlock integrated dictionary service, and while it isn't as good as a local service (in that I need to sometimes work offline), in other ways it is better (dynamic updates, multiple services, etc.).
- Even the dreaded Dock has its upsides: it is more visual feedback, and can manage some things better than window shade, the application menu, or the Apple menu that it tries to replace. I think if, or more likely when, Apple tweaks it a bit more, it will be much better. Just break out the separate context (let me have different ones that I can move where I want and work with tabbed windows / popup folders), give me a little visual cue (tab) when they're hidden so I know where it is, and lock the trash can in the lower right corner as a separate dock, and I'd be happy. Hmmm, wait a minute, that was Copland with a popup folder auto finding running items; who says you can't learn from others?
UNIX just works, many, many people know it, and write software for it, and have been for decades. It has entrenched markets, and very loyal users and programmers, and a whole lot of code. It succeeded because it is cheap: but success is success. People will borrow from UNIX, and people will bet on UNIX. Many of them would not have bet on Mac, without the UNIX foundation. This is significant; Apple can break into new markets because they are a UNIX, and they can drive technologies in ways they couldn’t before. It wasn’t that what they did before wasn’t as good (in some ways it was better, in other it was worse), but the biggest issue is not technical merit but marketing perception. Apple had proven unable to generate as much momentum on their own, so at least they can ride on the snowball that is UNIX. UNIX means opportunities for Macs and the Mac market that it wouldn’t otherwise have. This alone makes OSX such a huge leap forward over OS 9 that it is not funny.
While there are things that drive me nuts (negatively) about OS X, even those things that make me less productive (the Dock), often helps some other people and others even like. So I can’t call them complete losses, at worst they just aren’t quite as good for me. And I’m the only user that counts, right?
I like many of the new features. I really like the openness and trying to play well with others and make standards. I really like that I’m seeing continual evolution and progress: Apple is challenging themselves to be better. I love that they are extending what an operating system does, and offering better services (like calendar and synching). Apple could be more open in some areas (they do the closed API thing too much), but they are also more open in many other ways than they used to be. Most of all, they are choosing which standards to follow wisely and following through with those decisions much quicker (can you say LDAP, ZeroConf, XML, and so on)?
There's a better atmosphere in the organization (more or less). And while their much more pragmatic (and Microsoftian attitude) of “ship it now, fix it later” sometimes grates (and effects quality), it is probably better than the old, “Hang on to it until it is perfect, and irrelevant” attitude of the old Apple. But all the other little stuff pales in comparison to the new markets stuff; markets and market perceptions matter more than technology, more than good interface, more than anything. If you are a healthy market, they will come. Things are healthier. This is good. Add that to a few changes in attitude and you have some serious wins.
 Format for Printing  Mail
|
|
|
|
