Yes, but from what I can tell jailbreaking requires you to install new firmware on the phone. This isn't something any app on the phone can do. And there's no way developer deployed software can do this either. Please correct me if I'm wrong.

So I don't think your argument against DRM holds. DRM on a non-jailbroken phone seems to do a reasonably good job (especially compared to alternatives) for preventing viruses.. and Apple's store model in combination with DRM does well at preventing trojans (largely because authors cannot be anonymous). However you are right that if a hole was discovered in an API that allows an application to change the firmware, then you may have an issue. But I don't think it's quite as useless as you seem to imply.

To be fair, I have no problems with a YWDTTWWWYTBWKB philosophy. As long as it's "YWDTTWWWYTBWKB because we actually do no best" (eg. Apple) and not "YWDTTWWWYTBWKB because we don't you to have a choice) (eg. Gnome) or "YWDTTWWWYT for no good reason at all"/"YWDTTWWWYT because that's the way it's always been" (eg. Microsoft). I'm not really much of a fan of the "YWDTTWWWYTBWKB because we've determined it statistically" (eg. Google) method either.

Is that really true? I haven't done any jailbreaking, but I assumed it required a bit of work on an iPhone. I mean you can't just visit a website or run an app from the app store that would break it, can you?

Even as a developer, could I create and deploy a program that would jailbreak my own phone?

I'm asking as I really don't know.

And if it does require doing something outside the ordinary (eg. running a special program on your desktop while syncing, or deploying a new firmware)... then to me DRM is still sufficient to prevent exploits, as there's still effectively no way a rogue app will be able to do much harm.

I really like that game idea you posted. I'd certainly play it!

ReplayTV came out a bit before Tivo (though not by much). I rather liked the ReplayTV interface, and they never tried doing anything stupid like putting ads in while I'm trying to skip them. But they were far worse at marketing than Tivo. I've never used a Tivo, so I can't comment on their UI. But from what I've heard from others, the ReplayTV did have a few features up on Tivo (eg. the various commercial skip modes actually worked well).

These day's I'm using AT&T U-Verse's DVR, even though my ReplayTV still works fine... largely because of the integration with U-Verse. And while I really miss some of the features of my ReplayTV, it's hard to beat Microsoft's business model... cheap and good-enough.

Despite how well a $100-200 DVR can be, the fact that cable companies are giving them away means it's now a commodity... Most people just won't be willing to pay that much for something they can get for free.

Source code is useless if you're locked from installing your custom OS on the device.

This is the traditional way cellphone companies have dealt with their phones. This exactly why the "openness" of Android would be it's own down-fall in this market.

Despite what anyone says, Apple's "walled garden" is one of the greatest innovations in cell phone operating systems in that they actually managed to prevent the carriers from having the control to be stupid.

I think you're missing the point. What's going on is people think that the tablet is just about the hardware. So when they say "making a good tablet isn't good enough to sell a good tablet" they're talking about the hardware. The key ingredient no one seems to be talking about that really made the iPad was iOS and the developer infrastructure they had setup for it. Not only did they leverage the SDK from Mac to iPhone to iPad, but they also setup the whole distribution chain for 3rd party developers, which is what gave them a huge advantage in apps.

The combination of JITs and faster hardware has mitigated the performance issues of Java. But I do know that to this day, I *still* dread when I hit upon a website that causes my JVM to load. My Mac (and I believe Windows?) deals with this problem by pre-loading the JVM on startup.. And I believe the JVM is something like 500-700 MBs of memory.

If you run a Java app on a smaller embedded system (think i386-i586 class), you'll definitely notice the difference.

However, I'll grant that in many cases you may not care about that difference.

It's really sad, though, because Inferno was out and usable even back in 1995, and really did offer a small footprint (smaller than flash: http://slashdot.org/story/00/12/15/0428227/Inferno-Plugin-for-IE---An-OS-In-Your-Browser), run-anywhere JIT, etc. etc... it delivered on every one of Sun's promises for Java (which other than platform-independence, I don't think Java has yet delivered quite as nicely)... it just couldn't beat Sun's hype machine.

I agree part of the problem is cultural... it can be summed up in this nice little quip:

"When Java is your hammer, every problem looks like a thumb."

Java makes it really easy to use things like XML... But the language does also make it really difficult to do things at a binary level. This inherently makes all the Java people make everything use XML when it doesn't need it... And that's what gets you started down the slipper slope of layers and layers of additional complexity.

Yes, I'm perhaps overgeneralizing. But that's my 10,000 ft view.

Depends if it's an African or European program.

A slightly different topic, but one that I've been curious about regarding header files... Some header files are not merely interfaces. The kernel's linked list approach, for example, makes clever use of macros with no supporting .c file.

If I use that in my code, am I bound to the GPL? It's "technically" not linking of any sort, since it's all macro expansion... But there is logic there, and I could certainly see it being worth copyrighting.

I end up just re-writing it myself, but the end result looks very very similar... there's only so many ways one would commonly implement linked lists in C. And even fewer ways as macros.

Exactly. The problem is that information is a form of power. So, then, people with the right secrets will always be able to exert form of control over those without the secrets.

In order for a truly honest society to work, we need to have a Borg-like collective connection between everyone, where it simply wasn't a choice to lie or hide information.

I for one think this would be a good thing, contrary to how it was portrayed in Star Trek.

For a commercial organization, it is rather innovative.

It's clear from Apple's movements that they've pulled out of the Enterprise market, and justifiably so as you described very well. And while lots of people on Slashdot may want Apple to support the Enterprise, I don't think this move has much to do with that market.

I'm surprised no one else caught this (so maybe I'm way off base), but I suspect the main reason they're now incorporating OSX Server into Lion now is because of the iPad. As you can see in their feature list, they've streamlined file sharing with the iPad. I suspect they're anticipating many households (and businesses) to have multiple iPads but needing better synchronization/sharing. So while they don't necessarily want to support the enterprise, they do need the enterprise-like features to manage multiple users, file/resource sharing, etc. in a more convenient format.

Better synchronization/integration with OSX also means iPads will lend themselves to work better in an OSX environment, perhaps adding to the list of reasons for people to convert from Windows.

The way this *may* tie in with the Enterprise space is also I think due to iPads for the same reason. Perhaps Apple doesn't see the need for a rack system, but wants to provide an Enterprise way of managing large numbers of iPads. In this case, server processing is probably(?) not huge and a Mac mini may be sufficient. This doesn't sound like as strong an argument to me, but it's possible there's some motivation there.