Today is a good day to code

Mac OS X 10.4 – Review

Posted: December 31st, 1969 | Author: | Filed under: Uncategorized | No Comments »

Mac OS X 10.4 “Tiger”

Picture of Irv Owens Web DeveloperSince I'm a Mac user, I feel that it is my duty to do a quick and dirty review of “Tiger.” After all, there are so many sites that seem to be content to remain in Apple's good graces, I haven't found many that have been all that objective. I am not an ADC member, so I really don't have anything to lose by calling Apple on any of their exaggerations or lies. The Ars Technica review of “Tiger” was very good, and I don't want to try to redo what they or anyone else has done, so I'm going to proceed in discussing the well known points of “Tiger” from a user's point of view, then I'll discuss some of the lesser known changes and updates of “Tiger,” and finally I'd like to discuss some of the ramifications of what Apple has done with some of the new APIs they have added to their OS.

I, unlike many, eschewed the advice of most of the Apple news / discussion sites and did a simple upgrade from Mac OS X 10.3.9. I didn't feel that I should have to erase and restore everything for a point upgrade. If it was the case that I always had to do an archive and install, or a wipe and install, I may as well continue using Windows. My original reasons for switching to the Macintosh back in 2001 were very simple:

  • Spending too much time on maintenance
  • Tired of spending money on fighting spyware / malware
  • Insistance that with a decent GUI Linux would be the OS of choice
  • Free OS X developer tools
  • Macs are cool

So why would I do things like move all my files to another drive, and then move them back? Fortunately for me, Mac OS X 10.4 installed beautifully. It took about 30 minutes on my G5, and about an hour on my iBook. Both systems then took about 4 hours to index.
Spotlight
Let's start by talking about the feature that has gotten the most attention due to our search crazy eSociety, take a look at Google's stock price. This feature was not that important to me in the larger context of things. The individual applications I use, iTunes, iLife, Office, etc… all have decent searching functions built into them. My folders and files, while spread across two file servers and over 200 GB of internal storage follow an intricately balanced system devised by me over about 20 years of computer use. Suffice it to say I never really had trouble finding files on my computer. At least files that mattered to me. Still, I find Spotlight to be cool not for what it is today, but for what it could be in the future. Metadata is great, but you still have to enter it. Apple has just opened the door for third party application developers to go wild with helper applications for OS X. Imagine, a small application that will go out through the web and get lyrics for each and every one of the 5 to 10 thousand songs in your iTunes library and put them into metadata for their respective song. Think about the possiblities for images. Another third party application could ask you to load up images of your family members on black backgrounds so that it could generate a clean algorithm for matching their faces to pictures. It could then generate metadata for images putting in who is in the pictures, it could even find common patterns like text in images and add that to the index, like the sign for the camp that your son went to last summer. While this would be technically very difficult, it is not beyond the possible.
Spotlight is more useful as a framework for future development than something that will get massive use today, unless you have many hours that you can devote to typing in your own metadata for each of your files, or copying and pasting it from various internet websites. Without that metadata it is only marginally better than the search in OS X 10.3, but with the metadata, it is of great use to the vast body of computer users who don't have an organization scheme for their computers and do not follow naming conventions for their files.
Dashboard
I don't know if I am the only one who is super psyched about this, but Dashboard is probably one of the most important innovations of the last few years. I know all about the Apple stole it, but wait they didn't steal it, they came up with it, etc… I'm talking about dashboard as a concept. What Apple has effectively done is to create a way to allow the vast number of web developers in the world to create applications that can run in stand-alone fashion on Macs without learning anything new. The Dashboard framwork is extremely impressive. It allows developers to write widets using CSS and XHTML, in a standards compliant mode, that can execute shell scripts to perform system level tasks. If this sounds a bit like ActiveX to you, then you get the door prize. Apple is again kicking Microsoft in the chops by taking an idea they came up with and executing it in a much better fashion, embracing the existing developer community. Microsoft has to remember, its about the developers stupid! No OS is going to survive if people have to learn a new programming language with each revision, ie VB 2005. Also, if it costs thousands of dollars to start writing software for a platform most developers won't touch it, unless some kind corporation will allow them to muck around with their source code and servers.
Dashboard is a stroke of genious, and hopefully the hacker community will stay out of it and allow it to thrive and be what ActiveX should have been. I think that as long as billy bob and ray ray don't download widgets that are from “Gator” as they are prone to do then things should proceed nicely. Again, there is nothing extremely useful for Dashboard right now, but soon there will be.
iChatAV
I was reluctant to cover iChatAV's improvements becaus most of the mac audience really can't take advantage of it. The Ars Technica article does a good job of describing the ins and outs of this enhancement, but the more salient points are that 1 to 1 video conferencing only works with either a fast dual G4 to any G5, or G5 to G5 and requires 100kbps. Otherwise, it falls back to H.263 and the framerate drops to 15fps. This is somewhat understandable as it is pretty tough to compress video on the fly, but it shouldn't be worse than it was with 10.3. I think Apple needs to fix this.
To do the super cool three-way video conference a dual G5 needs to host it, I would hate to be in the same room with that G5, especially if it were one of the non-liquid cooled models, talk about loud, but the hardware requirements are understandable again because coordinating packets for four computers with non-consistent geographic locations and network connections is not a task for the faint of heart. But the 1Mbps upstream is going to be tough for the people who really want to use it, families. I don't blame Apple for this though, it isn't their fault that we have super litigious broadband ISPs in the US who don't really want to see technology advance. There is no reason we shouldn't have synchronous speeds over cable, and that it should be reasonably cheap. They have the money to deploy an advanced fiber network, and if that is too expensive then deploy wireless.
Developer Tools 2.0
Xcode is much stronger in this release with updated documentation and the ability to have it automatically flowchart methods. The ability to distribute builds across multiple computers using Apple's Xgrid technology, with the addition of a new GCC compiler with more automatic G5 and AltiVec optimization tools makes it completely indespensible. I thought the original Xcode tools were really sweet, and didn't know how Apple could make the tools even better, but they have. Apple is obviously a company that can learn from its mistakes and avoid them in the future, which is what most corporations fail at miserably. Many good companies have withered and died because they failed to change, and Apple realized after the bad OS 9 days that without a strong developer community they would be doomed. Objective-C is still really hard to learn, but now Java developers have as much to love about OS X as Obj-C developers do. Apple has done a great job with that.
General
Apple fixed a bunch of stuff that was wrong with 10.3, and enhanced some of the applications that you didn't know needed enhancement. The new system profiler does a much better job of giving users relevant information, where the old one, ie graphics cards being called by their chipsets, didn't provide much help. Now it gives you a detailed description of the graphics card, in plain english, and the enabled technologies ie Core Graphics, Quartz Extreme. They have greatly improved the speed of Safari and Flash within Safari. The overall OS is much more fluid and responsive than 10.3. Networking with PCs is much much improved as is wireless speeds. 10.3.9 for example would dump the finder whenever I was doing a large file transfer from Mac to Mac over wireless. My G5 would cause my PC to freeze for long periods of time while copying files across. That doesn't seem to happen any more with 10.4. Window resizing is finally smoother with OS X's enhanced line drawing abilities. It just feels really solid now.
The Future
Apple has left Quartz 2D Extreme turned off in Tiger's .0 release. I can only assume that this is because of some bugs found late in development. Users who have installed the developers tools can turn it back on by running Quartz Debugger found under the developer>Applications>Performance folder. I have found that window resizing and text scrolling get an incredible boost on my single 1.6 GHz G5. Of course it doesn't work on older PowerBooks and even recent iBooks because they don't have fast enough video cards and not enough video memory, ie Radeon Mobility 9200. I have yet to notice any artifacts, or bugs with it. I can only hope it will get tuned up over the next few months and enabled in the .1 release. Another really cool feature is the ability to change the dpi of the screen to make everything smaller on a 1024 x 768 screen for example. This is also available by the Quartz debugger, but appears to have a way to go before it will be ready for general release. It is promising though that someday I will be able to get a boost in screen real estate at the expense of my eyes.
It seems that Apple has done a lot of work with frameworks, but there isn't much to see in 10.4.0. At least not much for the average Joe user to see. It really isn't that much different than 10.3 as far as most people are concerned. Spotlight will be cool, but is it really worth $129.00 to the masses, no not really. I bought it because I wanted the developer tools to play with, and I am into frameworks and programming, but nothing about the new OS is so cool that I just have to have it. I know that Apple will start releasing updates that have cool functionality that will only work with “Tiger,” since that's Apple's modus operandi, and this will force its user base to upgrade, but cmon' is it really that much better than 10.3. It will probably get better at 10.4.3, but until then I'd say that the majority of users could probably wait.
iWork
This was almost not worth mentioning, but I thought I had to say something about it. With most if not all OS X 10.4 purchases, users are treated to a 30 day trial of the iWork suite. What is this a joke? This isn't a suite, Pages is like Microsoft Publisher but with Apple's world renowned marketing engine thrown at it, and Keynote is OK, but I'm not dropping PowerPoint anytime soon. There is no spreadsheet application at all, if this is supposed to replace AppleWorks, they have a very long way to go. I really have no use for it. The last Word upgrade was pretty good and Apple is not going to dethrone it with Pages, and the functionality of the rest of the suite still goes unmatched. The closest thing to it is OpenOffice. Seriously, I wouldn't have even released iWork until I had a spreadsheet application that was Excel caliber, an integrated email / PIM application, and some kind of database application that was based on MySQL. What was the point? Their sales are awful because most of the users who need this kind of stuff already have either Office v.X or Office 2004, both of which are better. I was really hopeful for this suite, thinking that it would be a better looking OpenOffice, but I was greatly disappointed. Hopefully by iWork 2006 it will be at least a useful low cost replacement for Office, but I still see no reason to upgrade from AppleWorks 6 if you don't have Office.