More of my sites

WinInfo Daily News
SuperSite for Windows
Windows IT Pro Magazine
Connected Home
Thurrott Dot Com
Windows Weekly at TWIT

About this site

For six years, the Internet Nexus served as my technology blog, but I've since started blogging at the SuperSite Blog instead. If you're looking for the blog, please head there. --Paul

Thursday, April 24, 2003

Launch it, launch it good
I'm in San Francicso for the Windows Server 2003 launch, and so far it's been a weird trip. I saw a car totally engulfed in flames at the airport, and people were just standing around taking pictures. There was a mouse running around in the gate area, freaking out the women (and the guy next to me with sandals on). At the connecting point in Chicago Midway, the plane slammed into the ground so hard that the flight attendents went around asking if everyone was alrgiht; if the guy next to me hadn't have been wearing his seatbelt, he would have catapulted into the exit row space in front of us. Good times.

[ Posted at 10:43 AM | Permalink ]

Tuesday, April 22, 2003

w.bloggar test
Just a short test posting using the Windows-based blogging application w.bloggar.

[ Posted at 10:00 PM | Permalink ]

Monday, April 21, 2003

Firebird vs. Firebird, the plot thickens
I've just spent a bunch of time reading all the blog posts from Mozillians about the Firebird naming controversy. The Mozillians are wrong, dead wrong, and their argument boils down, almost universally, to "AOL's lawyers looked into it and it came up fine." It's not fine, but what really rankles me is the unprofessional and petty way all the Mozillians are belittlng the guys from Firebird DB (most of them have even started putting little TM symbols next to Firebird every time it appears in text ... blanch). I hope Mozilla.org is forced to change the name and, more importantly, apologize. I have never seen a bigger group of sissies in my life, and I deal with Mac and Linux users regularly. Shame on all of you.

[ Posted at 8:36 PM | Permalink ]

Some thoughts about OS X "Panther"
The next major release of Mac OS X, dubbed "Panther" will launch this fall, but Apple is previewing a beta build for WWDC antendees this spring. In the meantime, we have this interesting "eWeak" report, which discusses a few Panther features, some for the first time:

User at the Center - A new interface paradigm that purports to make it simpler for individual users to personalize their computing experience and to move seamlessly among Macs and other devices. As "eWeak" notes, this feature is designed to copy the task-based UI in Windows XP and, apparently, the improvements to that UI that will appear in Longhorn, due in 2004-5.

Mobile computing features - Users will reportedly be able to take home directories they've saved on peripherals and networks and use them for file access or securely log into a Mac running Panther. Sources said the feature will let users synchronize their home directories with mobile devices or log onto a system via Apple's .Mac service. Apple's synchronization features are currently unparalleled in the Windows world, and if this happens, Apple will be a step ahead yet again. As someone who uses multiple PCs and Macs, and needs to seamlessly move from machine to machine, this feature sounds very exciting to me.

Fast user switching - They wont call it that, but Apple is reportedly stealing the idea for Fast user switching from Windows XP. No doubt, Apple fanatics will claim Apple invented the technology but, in typical fashion, I expect Steve Jobs to add at least one killer feature that differentiates it from FUS, so he can at least claim they got it right. (For example, in XP, when you switch users, it drops the Internet connection if you're on dial-up).

Piles - Patented by Apple two years ago, news of this UI feature began popping up on Mac Web sites a week or two ago. Basically, piles are collections of documents represented graphically as stacks. Users can browse the "piled" documents dynamically by pointing at them with the cursor; and you can divide them into subpiles based on each document's content. There is nothing like this on Windows, of course, but Longhorn will include a database-based file system called WinFS that will help users keep track of "their stuff." I suspect piles is a small subset of that functionality.

Small performance improvements - Frankly, OS X needs massive performance improvements, but that will require a better performing underlying architecture. And that brings us nicely to ...

64-bit support - Panther will reportedly support the 64-bit IBM PowerPC 970, though it will of course continue to support G3 and G4 chips as well. It's unclear how Apple will market 970-based machines, but I'd hate to see the company relegate them to the server line and then phase them down over time. They need better performance stats, and they need them now. If existing OS X apps can run on the 970 without change, all the better.

The biggest thing Apple has going for it, interestingly, is Microsoft's suddenly glacial development time. It used to be that Microsoft would pop out Windows releases every 12-18 months, but Longhorn is going to be a Longwait as well, and Apple has an interesting opening now to make massive improvements quickly, if they can. I'm not sure they can do much to gain back market share, but I love watching them try.

Next up for Panther: The WWDC. Should be interesting.

[ Posted at 5:24 PM | Permalink ]

Switch off
Amen.

[ Posted at 3:37 PM | Permalink ]

You know, Dave Winer is a jackass
But then, I've been saying that for years. However, it's always nice to have validation. My problem with Winer and his collective band of followers is that he leads even more of a back-slapping, circle-jerk than Apple fans do. It's weird, and should be embarassing to everyone involved. It certainly makes me cringe from a distance.

[ Posted at 3:28 PM | Permalink ]

Non-information
I was working through the resources for Beekman 6e today and came across a definition that applies to the Gene Steinberg phenomenon I describe below. It's called non-information, and it's defined by visionary Richard Saul Wurman as "stuff that doesn't inform." Perfect.

[ Posted at 12:15 PM | Permalink ]

Why does he have to make it so easy?
Gene Steinberg is one of those guys that writes about nothing. But he produces such pro-Mac drivel that you'd think he was a David Pogue clone. Thankfully he writes largely for backwater publications (where he can do little damage), his own (lame) Web site (his version of an unfunny Seinfeld--it's about ... nothing!), and he's the author of the most useless Mac book I've ever purchased (OS X Little Black Book). Here's how Steinberg comes up with a story. He browses all of the Mac news and rumors sites and then he writes an article discussing what other people are taking the time to report. He doesn't actually write about "anything" per se, and he's certainly not helping anyone. He's just ... summarizing what he reads. Useless? Yep. Funny? Yeah, but not on purpose.

Today's Steinbergian utterance is a good example of the unpurposefully funny. Gene, apparently, has noticed that many people are complaining about bugs in Mac OS X "Jaguar," which is of course impossible because a) Gene has never had any problems with the Mac, and 2) Apple is perfect. Let's dissect his discussion of this phenomenon, and, as we do, imagine how this conversation might go if he was writing about Windows, and not Mac OS X. Actually, there's no reason to imagine anything, as he starts off with this claptrap: "I want to make it clear that I think Apple Computer has done a simply tremendous job with Mac OS X. Consider how long it takes Microsoft to come up with a usable release (all right, some say they are never usable) with a staff much larger than Apple's. Apple's operating system guru, Avie Tevanian, and his crew have managed to do the impossible, and that is to make Unix warm and fuzzy for the average PC user. That's no easy accomplishment." In other words, Mac OS 10.2.5, which is the fifth minor update to Jaguar, is somehow a "tremendous release," while Windows XP (or presumably XP SP1) is ... what? Not an accomplishment? Didn't Microsoft make NT "warm and fuzzy" for the average PC users? Isn't XP used by over 100 million people every day??

OK, whatever. Here are the ways in which Gene excuses problems with Jaguar (I've added occassional emphasis for purposes of humor):

"Nobody is perfect, and so it's understandable that problems will occur during the development path."

"There are so many possible Mac hardware configurations that nobody's quality control lab or even a wide base of beta testers can possibly cover every possibility, or even a reasonably large number."

"There are quite a number of USB hubs out there, and the USB hardware on Macs has undergone various changes from its first appearance on the original bondi blue iMac. I don't believe we can expect Apple to purchase every make and model of USB hub available and try each, in turn, with every Mac that's compatible with Mac OS X. The possible number of combinations would be mind-boggling."

"It's quite possible that, once Apple nails it down (and it will), it'll be traced to a conflict with a specific product line or chipset ... I don't expect it'll take terribly long to solve, and thus we might see a 10.2.6 update before long, or just a specific patch to fix this specific bug.

"Some Mac users that can't wait will just be using Jaguar's Archive & Install feature to set up their operating system all over again, and return to 10.2.4. With that release, you'll have to worry about other bugs, such as the date reverting to 1969 or 1970, but at least you may not encounter a kernel panic."

"It's possible, of course, that the [battery life] problem [currently plaguing iBook and PowerBook users] is coincidental. Maybe the batteries on these laptops were due to fail anyway [!!!!].

To summarize, though I do believe Gene's assesment stands as a critical work of research, I present the following:

Nobody is perfect.

There are so many Mac configurations and different USB hubs out there that Apple couldn't possibly take the time to test its operating system against all this hardware.

Apple will fix the problem, don't doubt it.

OS X users can use the intuitive and simple "Archive & Install" feature to reinstall an older version of the operating system. LOL.

Those battery problems don't affect Gene's $3500 laptop, so they don't exist. And really, weren't you ready to buy a replacement battery anyway, you skinflint? Come to think of it, why haven't you just purchased a $3500 Powerbook-17?? This is the year of the notebook, after all.

Can you imagine anyone--even a Microsoft PR flack--cutting the software giant this much slack for problems with Windows XP? I didn't think so.

[ Posted at 9:13 AM | Permalink ]

NPR drops Apple QuickTime
NPR writes, "For several years, npr.org has provided its 24-hour program stream ... in Apple QuickTime format, as well as in the Real and Windows Media formats. All of the ... Apple QuickTime streaming has occurred under agreements with Apple QuickTime. We have been renegotiating these agreements with Apple QuickTime for a number of months but we have failed to reach mutually acceptable terms. As a result, we will cease offering any NPR content in Apple QuickTime format effective today (April 9, 2003)." Apple has come under a lot of scrutiny lately for bizarre new licensing practices, and NPR seems to be just the latest casulty.

[ Posted at 9:04 AM | Permalink ]

Sunday, April 20, 2003

Five years of Mozilla
Speaking of Mozilla.org, LWN recently published an article called Five years of Mozilla, an interesting, if-overly positive look at the first half-decade of Mozilla's life. I wish I could be equally positive, but I've watched Code Rush, I've read every book there is about Netscape, and I've been covering Netscape/Mozilla news at my day job for years, so I know the truth. So, the short story goes like this: Mark Andreessen did not invent Mosaic or even write most of it. When Marc joined up with Jim Clark, he was unable to come up with a compelling new business, so they eventually settled on re-doing Mosaic (this was about their 7th business plan as I recall), so they hired all the Mosaic developers and Netscape (eventually) was born (they tried to use the name Mosaic, but were sued). The code Netscape wrote was complete crap, so when they released the source code to Communicator as open source in 1998, excited developers at the recently established Mozilla.org quickly realized they'd have to restart the project from scratch, which they did in late 1998. A short four years later (ahem), they released the first version of their browser suite, Mozilla 1.0, a bloated but standards-friendly browser with an excellent rendering engine that quickly grabbed less than 0.1 percent of the market (cough). That brings us to the present. Less than a month ago, Mozilla.org realized it had to dramatically change course, decouple all the applications from the suite and, yes, start all over again. This move, while technically sound, should have come four years ago and will likely garner them another 0.1 percent of the market. Woo!

Sounds pessimistic? I guess so, but it's also realistic. The sad truth, by the way, is I use Mozilla on the PC, where its performance problems are hidden by the PC's superior architecture and speed. But I've switched to the excellent and speedy Safari browser on Mac OS X (from Chimera, which is based on Mozilla's Gecko rendering engine) for performance reasons. I still use Mozilla on Linux as well, though I'm interested in Epiphany. Why don't I use IE? It's buggy, insecure, and doesn't offer tabs or pop-up ad blocking, two crucial features that virtually every other browser in the world offer.

Anyway, my point is that we can cheer Mozilla on all we want, but they've accomplished a startlingly small amount after five long years of work. In fact, Mozilla and its variants (Netscape, AOL, etc.) have done so poorly, that Microsoft's Web browser development has come to a virtual stop in those same five years, and Internet Explorer (IE) still controls over 95 percent of the market. And that, gentle reader, is Mozilla's real legacy: Defeat after defeat after defeat.

[ Posted at 3:37 PM | Permalink ]

Firebird flap, first Thunderbird release
Mozilla.org finally decided on the name Firebird, after discovering that the original name of its standalone browser project, Phoenix, was already taken. There's just one problem: Firebird is already taken too. So while Firebird might be an excellent name for this project, it seems that Mozilla.org might have to change it yet again. So far, there's been no word on whether they are even considering this.

In related news, the first Thunderbird-branded stand-alone Mail/News build is now available, but only for Windows. Looking good so far.

[ Posted at 3:28 PM | Permalink ]

Abusing trust
One thing I take very seriously in my day job is that readers can trust what I write. I try to be as honest and educated as possible, and one of the things I do that most tech journalists don't is use non-Windows systems, such as Mac OS X and Linux, regularly (heck, I'm writing a book entirely in Mac OS X right now, actually). This gives me, I think, a unique--dare I say, "experienced"--perspective that many people lack. As part of this experience, I spend an ungodly amount of time checking up on Mac and Linux (and Mozilla) news and rumors sites. Most of the Mac sites stink: They're full of tunnel-visioned trolls whose opinions of Windows were formed a decade ago and never updated. Most of the Linux sites also fall into easily-labeled categories, too. Virtually every Linux distribution "review" out there, for example, simply describes (usually in breathtaking detail) how simple the installation was and how everything just worked (usually after entering some bizarre collection of command line commands, of course), and then... that's it (Heck I can read those kinds of Linux reviews on this site. Oh wait). There's very little in-depth analysis in the Linux world, for some reason, which is exactly the opposite of what you'd expect. I have been heartened, recently, however, by stories like this and this that demonstrate that even the Linux community, finally, is waking up to the fact that their favorite open source project isn't for everyone.

And then there's this misguided soul. There is nothing as alarming as reading that some Linux fanactic (whose email handle is penguinrox, naturally) has decided to "help" his neighbors by getting them going on Debian Linux (virtually the hardest Linux distribution to install and use, incidentally) and then find open source equivalents of all the applications they actually need to run. This is wrong for so many reasons, it's hard to understand why I even have to explain it, but after a couple of email exchanges with the bozo, it's clear that not only is he living in a fantasy land, he's interested in taking down others with him. This is both sad and alarming.

Like any sane person, I'll be praying for this idiot's neighbors. They're in for a world of pain. I just wish I could be there when they finally figure him out and storm his home in a Frankenstein-esque rage. He would be strung up by his toes in some countries for hurting people like this.

[ Posted at 3:15 PM | Permalink ]

Giving Unreal 2 another chance
I buy virtually all the first-person shooters that come out for the PC, and though I still prefer Quake III Arena over everything that's come out since (I recently completed both Quake III and Quake III Team Arena in Nightmare mode, by the way), I've had some good times with various games, such as Unreal Tournament 2003 and Medal of Honor. While in New York a month and a half ago, I picked up Unreal 2, and after trying it out briefly, I kind of forgot about it. Unreal 2 really stretches the limits of today's top-of-the-line hardware, and on my system (a P4 1.8 GHz, 1 GB RAM, 128 MB GeForce 4 Ti 4200), its barely acceptable at 800 x 600 (sort of leisurely in an iMac kind of way). The game is definitely gorgeous, and arguably the most graphically sophisticated 3D video game ever created, but I just found it boring and slow-paced, and hated the fact that non-player character and enemy "skins" didn't look right on my system. The other day, however, having finished the aforementioned Quake titles in the hardest possible mode, I was looking for something different, so I loaded up Unreal 2 again. And after a few days of playing it... hmm.

First of all, the game is beautiful. It's got a plot, which slows things done, frankly. I fixed the skinning issue by loading the latest NVIDIA drivers on XP (I was positive this wasn't going to work, and had actually written to Epic Games for suggestions, but hey, it worked). I've been playing through the game and it's not bad. Some of the levels are quite inventive. Some you just have to get through. But I've come away rather impressed, overall. It's still not as good, in my mind, as an all-out shooter, but it will pass some time while I wait for DOOM III. And I'll definitely finish it, what the heck.

[ Posted at 3:07 PM | Permalink ]

Printing to a Windows shared printer from Mac OS X
Ever since I got my iBook back in mid-2001, I've been trying to figure out how to print to my HP LaserJet 5P, which is shared on my workgroup from my desktop computer, which runs Windows XP. Over the ensuing months, Apple has made it easier to work with Windows-based shares, thanks to its integrated Samba technologies, but printing just never seemed to work. Last week, when I installed Red Hat Linux 9, however, I noticed that printing to the XP-based printer worked flawlessly, the first time, and it got me thinking. If I can do this from Linux, it's gotta work from Mac OS X, as they both support CUPS printing now. And, it turns out, it does work, and I got Windows shared printing working today from the iMac. But man, is it ever difficult, and very reminiscent of the way accessing Windows shares used to (sort of) work under the original Mac OS X version. The problem is that I'm not sure exactly what I did to make it work, so I'll have to test this again using the iBook. But I believe it involved at least some of the following steps: 1) install UNIX printer services on XP. 2) Create a new printer share on Windows using LPR and not auto-detect LPT1. 3) Install some crazy Ghostscript and EPS software on Mac OS X. 4) Make sure Windows sharing is on, and that you're on the right workgroup (which, inexplicably, requires two completely different applications in OS X; typical). 5) Use the correct LPR over IP connection to connect to the new share. 6) Print. Or something like that. I'm going to test this soon and see if I can't do a better job of documenting this, but here are the resources I used to get it to work. Note that none of these, by itself, actually worked, but that a combination of these resources caused me to finally get printing working.

1. How to Print to Windows via SAMBA and CUPS
2. Printing to Windows from MAC OSX 10.2 (Jaguar) using CUPS and SAMBA
3. How to Use a Printer Attached to a Windows XP Computer in Mac OS X

And if you're trying this at home, good luck!

[ Posted at 3:00 PM | Permalink ]