Oren’s Weblog

I’m working from a Starbuck’s via wireless this morning while my car is being serviced. I’ve paid my $10 to t-Mobile for a day’s worth of connectivity (this suburban neighborhood doesn’t have any place in walking distance offering free wifi, otherwise I wouldn’t be here).

I got connected ok and went off reading my email and trying to catch up, but the first time I tried sending an email I got an error message – then the fun began.

Like most other academic institutions, and, I assume, lots of other places, we now require people connecting from off-campus addresses to connect securely to an authenticated SMTP server for sending email. That way people can continue to use our mail servers, but it keeps them from being used as open mail relays by the various dark forces out there on the net.

But it doesn’t work from the t-Mobile wireless at Starbucks. It appears that they block SSL-encrypted connections on the usual mail-sending port (25). Why would they do that? It makes no sense whatsoever. And the caffeine coursing through my veins didn’t make me any calmer as I tried to figure this out…

Luckily, I could get into our unix machines just fine with a Kerberos-authenticated telnet connection and use good ol’ Pine just fine.

After a couple of exchanges with our fabulous support and engineering staffs, it turns out that I can get regular Mac mail working by switching to an alternate port (587) on our SMTP server.

It’s great that we can get this to work, but how would an average Starbuck’s coffee swilling computer user figure this out? They wouldn’t. Then they’d be madder than hell.

After all the brouhaha last week about Greg Jackson’s comments on Macs, I was particularly amuused to see this in John Gruber’s Daring Fireball blog post about the details of the ‘MP3Concept Trojan Horse’:

No one with any sense would ever claim that Macs are impervious to viruses, worms, or Trojan horses. Especially Trojans — which just about anyone with a 3-digit IQ could put together. E.g.:

Voilá — you just wrote an innocuous Trojan horse.

For years I’ve thought that granting patents on software is a bad idea – it stifles innovation and turns the development over to lawyers instead of technologists.

Now the EFF has announced The Patent Busting Project.

Patents traditionally only targeted large commercial companies,” said EFF Staff Attorney Jason Schultz. “Now bad patents are threatening non-profits, small businesses, and even individuals who use software and Internet technology.” These threats target non-commercial personal use, such as building a hobbyist website or streaming a wedding video to your friends.

The new EFF initiative seeks to document these threats and fight back against them. EFF has pledged to file “re-examination” requests with the United States Patent and Trademark Office (PTO), asking the agency to revoke patents that are having negative effects on Internet innovation and free expression.

Reference: Bitlaw’s History of Software Patents

It’s been a lively day around the old blog! Keep those cards and letters coming

MacDuff noted in a comment on my observation that I have become a Mac user during the last year:

Super! It was surely a difficult thought process to go through. I totally understand the hesitations that frustrated Windows users can experience when mulling The Big Switch. And yet, YOU DID IT! Why? These reasons may also be just some of the reasons why that choice should be recognized “out there”. Let’s face it; you most surely didn’t do it just for the pretty boxes Apple makes, right? There HAD TO BE some pretty damned compelling reasons to switch offered by Mac OS X and many of Apple’s other technological creations. So, why surrender to the status quo, when that status quo may not be the best solution. Is resistance truly futile?

First of all, I don’t deny that there should be choice and heterogeneity in computing – and I even frequently argue for interoperability based on open standards – that, after all, is what the Internet is all about! But there are lots of forces that push against supporting heterogeneous computing environments – it’s easier to support lots of complex interdependent functionality in single environments (Microsoft Exchange is the example that we keep coming up against lately), and it’s usually perceived to be cheaper to support a single environment than to account for multiple ways of doing things. I merely meant to observe that these forces are at work in the academic environment as well as in the coprorate world, and those of us who want to fly the banner of open, interoperable systems have a constantly uphill struggle, perhaps most like Sisyphus.

There are indeed reasons why I switched most (not all) of my computing to the Mac platform over the last year:

- Much like modern cars, I have found it increasingly difficult to understand how Windows operates as it has gained complexity with each successive version. Unlike modern cars, I find that Windows systems become erratic and unreliable as I use them over time. The interaction of these two factors makes it very difficult for me to troubleshoot problems with aging Windows systems. The usual answer to this is to reformat your hard disk and reinstall Windows from scratch. I’ve always found that to be a problematic approach to fixing things (just burn it to the ground and start over again), and with the current security scenario, you’re never sure if you’ve reinstalled all the patches.

- I like OS X’s stability. My desktop iMac frequently goes a month or more between reboots, and I can consistently put my PowerBook to sleep just by closing the lid, move to a different wireless network environment, open it up and have it work. That has never been the case for me with Windows systems.

- I do a fair amount of work from a plain-text terminal environment, and OS X supports that better than Windows, due to its unix underpinnings. The integration of tools for working with multiple environments is just better – kerberos works like it’s supposed to, for example.

- I can run standard unix server kinds of processes on my Macs without having to buy or install additional software. The ability to ssh between my Macs is worth a lot to me. In many ways I view OS X as being the “Unix for the rest of us” (that being those of us too lazy or time-deprived to take on the administration of a full-blown linux system).

- Last, but not least, I do indeed believe that OS X is more secure than Win XP, for all of the kinds of reasons frequently cited, most recently by Lisa Spangenberg here.

But that’s not to say that the Mac is perfect (just last night iMovie crashed on me right before, of course, I saved the work I had done for an hour of editing footage of my son’s t-ball team), or perfectly transparent (can anyone really explain to me where OS X keeps track of the associations between mime-types, file extensions, and applications?).

And I’m certainly no fan of Microsoft in many, many ways – see, for example, this post, or this one, or (perhaps most especially) this.

But isn’t it a strange world where you can’t figure out how to have a conversation with development teams at Apple, where the OS is based on open source, yet Microsoft developers are writing blogs?

Wow! My post on Greg Jackson’s remarks on Apple and his subsequently being asked to resign from Apple’s University Executive Forum, got picked up in Paul Thurrott’s Internet Nexus, and have generated lots of comments, some more thoughtful than others.

In my post, I stated that I thought that this incident didn’t paint Apple in a very favorable light, and I stand by that opinion. While I don’t know all the details of the UEF, it seems to me that its main purpose is to give Apple some insight into what’s happening in the world of IT in higher education, not to be marketing Apple to their institutions. While many of the specific points of Greg’s comments can be debated (as he himself admits, and which I’ll address in a minute), Greg is an experienced and highly respected and influential CIO and his opinion and input should be valued by Apple if they care about this market. As one of my bosses has observed, the best strategy (at least in academia) is to populate your committees at least partially with your smart and articulate opponents – if you just talk to your friends you will always be blindsided.

Despite the objections from the Mac faithful that have commented so far, there is a lot of truth in Greg’s comments. One of the issues facing us in the higher ed environment is how long we can maintain our long-held beliefs in open standards, support for heterogeneous desktop environments, and freedom of choice when the world we live in is overwhelmingly populated with Windows desktops – and it troubles me deeply to make that statement.

I should probably state that I am a Mac user myself – I switched my primary computing environment from Windows last year, and I like my work desktop 17-inch iMac and my new 15-inch Powerbook, and I’m about to part with a chunk of my very own hard earned cash to buy a 20-inch iMac for my home.

But here at the University of Washington, which is probably typical of large higher-ed institutions, we have seen the Macintosh installed base dwindle from what was probably around 50% of desktops a dozen years ago to something around 10-15% today. And while we have seen some resurgence of Macs in the last year, based primarily on the adoption of OS X, it is not yet enough to call it a direction-setting trend.

And there are lots of people in higher ed computing who still bear the scars from having followed Steve Jobs a dozen years ago into promoting NeXT as the coming thing, only to be abandoned after a very short while.

Now – some specific comments on Greg’s specific comments:

- I do think Greg misspoke when he claims Apple charges for security updates. The patches continue to be free. But at least originally Apple announced their intent to have security updates to Jaguar only available as part of a purchase of Panther for $129. They’ve sinced backed off this untenable position and are issuing Jaguar security patches freely.

- While I would argue, and indeed did in email to Greg yesterday, that OS X comes configured far more securely than Win XP has, at least up till now (there’s some hope that SP2 for XP will help), that doesn’t mean that OS X is in any way invulnerable – last week there was indeed an announcement of a discovered Trojan for OS X. I know – the threat was undoubtedly exaggerated, but it still points out that it’s perfectly possible to conceive of trojans and viruses for Macs, even if they haven’t been widely observed in the wild yet. As a point in fact, there is an ObjectiveC API to the OS X Address Book, and it should be relatively easy to code up a set of innocent looking scripts that would pull exactly the kind of mass email shenanigans we’ve seen so frequently lately.

- As someone who hires a fair number of computing support staff, I will vouch for the fact that it has become more difficult to find people who really know Mac technology. More and more we find people who know a lot about Windows, and people who really like to get their hands under the hoods tend to gravitate towards Linux, especially college students.

- As Greg notes, the iMacs and iBooks are fairly price-competitive with Windows boxes. Just last weekend I was comparing prices between the 20-inch iMac and a comparably equipped Dell machine, and they came out within about ten percent of each other (though that’s still a couple of hundred bucks), though the Dell ended up with lots more disk for the price. I’ll still go with the Mac for my house. But he’s right in that you can get a lot more raw horsepower (which admittedly tends not to be what Mac people value the most) for your money on Intel boxes than you can with G5s.