What I’m interested in in technologyPosted: May 11, 2004
So I promised I’d post about this a week or two ago. This post is going to be a combined rant against current developments in technology, where Apple isn’t and should be, and why Linux can’t compete yet in the desktop space. In short, it’s going to be a rant as to why I’m still using Windows and probably will be for some time to come. Blasphemous, I know! Prepare for a long post (if you’re one of those non-technical readers, here’s a good point to bail out 🙂 ).
Currently, I use Windows for 90% of my daily tasks (I have two Linux desktops, one at home and one at work). Most of the reasons I use Windows are because of ingrained corporate mandates on software and document formats, and are not related to whether other open source or Mac-available software are or are not “good-enough.” In fact, I think the majority of the Open Source apps for the daily tasks I perform have reached the “good-enough” phase, but I’m unfortunately locked into using file formats which are proprietary (which later in this post we’ll get to the one point I agree with Sun on, that open standards are far more important than Open Source). Here’s a quick list of why I haven’t blown away Windows on my machine for Linux or why I haven’t bought myself a Powerbook for daily use:
- Only important on the Linux front, as Office can be had for the Mac, but 100% Office file format compatibility is a must. I have to modify complex file formats with macros etc enabled because for some lame-brain reason people decided that Microsoft Word was a good way for people to submit forms rather than doing a nice web interface.
- Visio. Microsoft was smart to buy them and ensure they remained a Windows-only software package
- iTunes. Again not a problem on the Mac, but Rhythmbox isn’t iTunes, that’s for sure.
And the real kickers:
- Outlook/Exchange. I’ve considered moving to IMAP for email, but the integrated calendaring is pretty essential for me. Evolution offers a plugin for exchange ($70! what a rip) and I’ve heard iCal and Mail.app are offering Exchange compatibility, but neither are as good as Outlook.
- SQL Server Enterprise Management tools. Our billing vendor uses Microsoft SQL Server and FoxPro for their DB backend and development language, and having the Enterprise Management tools local to my machine saves a buttload of time. I could work around this by using RDP to access the individual servers to run queries, but still I really like having the tools locally.
- Nortel Contivity VPN Client. Netlock makes versions for other platforms, but frankly they suck. The MacOS version might be better than Linux, I don’t know, but the Linux version crashes anyone but RedHat’s kernel (possibly the reason everyone is decrying RedHat’s “fork” of the Linux kernel, which, unless your running binary-only kernel modules like Netlock’s, isn’t really a problem).
I’m sure I could probably come up with some more reasons, but frankly, there’s lots of small issues that, besides applications I can’t run. MacOS’s eye candy is appealing, but Apple has really fallen behind the times in terms of UI design. None of the applications they produce are consistent (and don’t even get me started on Brushed Metal), and the Finder is just appalling (Gnome’s new spatial browsing goes back to what I actually like about the Classic Finder). The Dock is the worst excuse for an Application Launcher/Task Manager I’ve used (I much preferred OS 9’s Application Menu). I know I’m just reiterating old complaints here, but I’d like to make it known that I don’t think Apple is innovating on the UI front. In fact, I think they could really take some notes from Microsoft on making their UI consistent throughout the OS and bundled applications.
Now, don’t get me wrong, there are things the Mac does well. However, it’s a dieing platform, and thusly, even after using the most recent Macs, I can’t see paying a premium for a platform that with every passing day becomes more and more irrelevant. For the Mac-addicts out there, this amounts to pure blasphemy, but take it from someone who’s owned a Mac for the last 14 years, I’ve seen the ups and downs of Apple Computer, and there’s nothing Apple has left to give us. I may be surprised, but as everyone knows they’ll never be more than a niche player and that niche is only going to get smaller. For a home media PC, which is what the Mac is gearing itself towards now, there is no compelling reason for me to buy it at a premium over a Windows Media Center PC. While Apple’s iDVD, iMovie, etc, are very easy to use, I don’t see them as that much better than the competition to warrant paying the premium for the hardware, especially since other than iTunes, none of the applications interest me much (I’m not really into Video at the moment).
Now that we’ve established that I don’t think Apple is innovating much, it’s time to start complaining about Linux. As Bill Joy says, re-implementing the same OS that has already been written is not innovation. As much as I applaud Richard Stallman and the GNU group’s goal of truly free (as in speech, not as in beer) software, and it has given us a great base from which to build an operating system, GNU programmers have continued along the lines of merely re-implementing the software which is already there. GNOME and KDE are just starting to innovate after spending the last 6 or 7 years of their existance merely trying to copy Microsoft’s UI. While this stands to ease the transition for existing Windows users, I don’t believe that’s necessarily the way to bring about change. People don’t switch platforms because they’re cheaper. You can buy a complete bundled Windows system for $500, and the OS is only perhaps $45 dollars of that at OEM level. Thusly, the OS is not the factor that’s going to bring PC prices down (they’re already about as cheap as they can get). The reason people would choose Linux over Windows is a compelling technological innovation, either in the way they use the computer or something the computer does for them that it doesn’t do on other platforms. Linux is a disruptive technology on the server side, not because of the technology but only because people are beginning to realize that it’s easier to standardize on one OS, because the OS for the majority of users or administrators is irrelevant. It’s the applications that count.
Currently, Linux is not offering innovations in the way I use my computer. This doesn’t detract from it’s use a server platform, where innovation is actually occurring. I think Linux is a vast improvement over Windows in terms of both administration and reliability, although Windows 2000/2003 is quite reliable as a server platform not withstanding its security problems. However, as a desktop platform, Linux still considerably behind both Apple and Microsoft in usability. They’re catching up, but it’s still hasn’t advanced beyond a playtoy for geeks at this point. It could be used in areas such as call centers or retail Point-Of-Sale, where there is generally only one application running, but that’s not innovation, as UNIX desktops have had a role there for quite some time (this is where SCO made it’s bread and butter before they simply became a litigation company). There are some exciting projects in the Linux world, such as Storage, but they’re probably still a year or two off.
So I’m not happy with alternative platforms to Windows for my corporate use. Obviously this would be much easier if the industry would come about and implement standards based document formats. I’ve always felt that if the software was better it wouldn’t matter if you could easily open documents in a competing product. If Microsoft’s software was that much better than StarOffice or OpenOffice.org, then they’d have nothing to worry about with having others be able to easily read their document formats. Office 2k3 has allowed users to save their documents in XML format, but it still stores large portions of the Office data as blobs which are still basically unreadable to third parties, not to mention they’ve patented the formats so they could potentially block users from reading the formats w/o a license. This is truly where I think Microsoft shows their monopolistic behaviors! I really wish Microsoft would compete in the market place instead of trying to force the marketplace into staying with them. They’d truly endear themselves more to the marketplace if they’d allow people to easily migrate away from their software, because they’d probably find that as long as their software continued to be the best that the users would stick with them or the ones that left would come back if their software was truly worth the dollars they charged. See, I’m not a tried-and-true Microsoft lover, although what I’m about to say will probably make it look like I’m smoking Bill’s pole.
The scary fact of the matter is, no one is innovating in technology right now, except Microsoft. That is blasphemy you say! However, I’m not alone in saying this. Miguel de Icaza, founder of Gnome, agrees with me. .NET, as confusing a marketing ploy as it has become, is truly innovative. It’s almost what one could consider a disruptive technology. Java had the potential to be what .NET is going to become, but for desktop applications it’s failed for years because of performance issues (I’ve been told this concern is largely elimated today, however Microsoft seems to have gotten it right on the first go-around). The Common Language Runtime allows developers to use whatever syntax (as long as it can be compiled to the CLR) to interact with the Frameworks that Microsoft has provided to interact with the operating system and thusly the user. Obviously, the value is in the Frameworks, as the languages are just a way to express the program in whomever’s preferred syntax. The Frameworks and APIs lock someone into a particular platform, as unless portability is a design goal from the outset, it’s very expensive to build and much more expensive to retrofit. The Mono project is working on re-implementing the Microsoft Frameworks and an Open Source implementation of the CLR which could potentially bring true cross platform development to both Linux and Windows. Another example of following, but one at this point which is probably necessary to remain competitive.
Longhorn stands to bring many more improvements to the way we use our computers and especially in the way in which applications are deployed. This is the first time Microsoft will be innovating on the desktop in years, since the years of 95 and NT, IMHO. WinFS stands to bring to the mainstream a meta-data based Filesystem. This would bring customizable attributes stored on the filesystem to allow all kinds of new possibilities for displaying data, and with the advent of improved search technology, would make where you store your files irrelevant. This stands to bring vast improvement to the way users use their computers, as while the Hierarchical Filesystem makes sense to you and I, many users find it incredibly confusing. File management, with today’s ideas and technologies, should really be a thing of the past. Search is the thing now. Obviously, this isn’t the first time someone has implemented this idea, BeOS and many other experimental OS’s have implemented this type of filesystem previously, but this will be the first time it will brought to the masses with the potential that everyone could be able to figure out how to use it. Where’s Apple on this front?
Also, and even though this stands to put more into vendor lock-in with Microsoft, Avalon and XAML present very cool deployment technologies for applications. I have to say I’ve only read the hype, but basically from the way I understand it, XAML and Avalon provide a method for describing user interfaces in XML (much like XUL, but it will probably not be incredibly slow) and providing the needed .NET code in bytecode form to execute proper actions when the interact with the interface described in the XML. This basically would allow a web browser to display applications which to the user would look and interact just like a local application. Basically what Java and ActiveX were supposed to provide but never quite could.
Obviously, all this remains to be seen as to whether Microsoft will deliver on its promises. It rarely does, and its never met a deadline it could meet feature-intact. However, with the $7 billion a year or whatever they spend on R&D, it’s going to be hard for others to compete.
So, now that I’ve said that and pissed off all my friends for blaspheming the beloved Apple and praising the hated Microsoft, let me put this in context. If someone can point to something that’s innovative coming out of the Open Source world that’s ready for production, or something from Apple that’s innovative outside of Eye-Candy and easy-to-use digital image cataloging and video editing software, then I’m all ears. I just don’t see it, and I’m really looking for someone to change the way I use my computer at the basic levels. I’m not talking voice recognition, but at least something that makes my daily life easier like a database-driven filesystem or an email program like Gmail that’s local but still doesn’t require me to spend time to cataloging my email. Search should be powerful enough I don’t need to think about that stuff anymore. Maybe I’m just selling myself on search, but I’m believing the hype around Google and the search business. I think Internet search is important, but I think the most important aspect of search at this point is improving the way we use our PCs, as I think the Internet search thing is approaching mature and commodity level. However, as much as people have tried, I don’t think PC search is there yet. Maybe in the next few years I’ll be happy.
Can someone please point me to something which will pique my interest and show me that there’s something interesting in technology? Can someone show me someone who’s innovating and implementing new ideas for production OSs (academics don’t count, it rarely is ever usable) outside of Microsoft? I’m ready and hopeful to be proven wrong.