Dear Snarky Slashdotters,

Apple's iTV is coming. It's coming with an all-in-one TV, no set top box, in the same vein as UbuntuTV. iTV is going to eat the brainz of consumers just like iPod and iPhone and iPad did. Futuristic HD videophone is coming to the tv, apps and app stores to the tv, voice recognition is coming to the tv, artificial intelligence is coming to the tv, home automation is coming to the tv...

You think your quaint little Linux OS is the bee's knees, right? It sure would be nice if more people actually used Linux on a hardware platform that hosted apps, right? It would be fantastic to see more mainstream devs learning about and targeting Linux for development, right? Maybe Linux-as-a-platform on TV's could get a little foothold before Apple iTV storms onto the scene.

Just this once, rather than being snarky, you could think all strategic and positive and glowy about Linux. Put aside the odd name Ubuntu and the ongoing forward/backward version binary compatibility challenges in Linux. Put aside all your gripes about the state of desktop Linux, a separate issue.

Strip away all the technical stuff and consider now only this question ---> Can we position Linux as a prominent platform in a big segment of an existing, but poorly served, consumer market?

See now how important this move by Canonical is to get Linux on the TV? You and I know there's nothing ghetto about Linux when it's deployed right. It's in tons of embedded systems as the OS, so most people have some exposure to Linux. But Linux itself isn't the platform for apps development on those devices, something running atop Linux is, like Java. I don't know what UbuntuTV will use as it's runtime platform for apps, but if it's C/C++, then that means we could draw more developers to the Linux platform and have them become proficient at building products for it.

Please try to see the vision here: this is something to get excited about and rally around!

Dave

Oh, you're one of THOSE guys. You complain too much. Java isn't as bad as you say even if there's a whiff of truthiness in most of your criticisms.

I argue, Mr. Grouch, that the crown jewels of Java are not to be found in the language. They're found in the JVM. Plain and simple: the JVM is the most banged on, battle hardened, security-first computer programs ever written. There is no more trustworthy binary in the world more than java.exe. You want an unrockable web server? Run Tomcat with the NIO thingy enabled on the latest Java 6 atop Linux with the firewall all ratcheted up. And don't proxy through Apache HTTPD if you don't have to - that's just one more security-as-afterthought, million moving parts binary with perennial remote exploits to worry about.

Why do languages "that target the JVM" actually make a difference? Because you get all that bitchin runtime robustness without the Java language baggage you just love love love to complain about. And while we're at it: Java Native Invocation (JNI) is considered harmful. Native code tainting the JVM? You better have a darned important exotic business requirement to bring new native code into the equation. As far as I'm concerned, you're crazy as a loon to write native code these days unless you're doing embedded systems or device driver development. I'm wondering why you would suggest something so reckless! Care to elaborate?

I know the JVM's design makes life a real drag sometimes. It feels like a trusty slingshot that's been upgraded into a WMD. But, I could care less about aesthetics or the angst about Java that computer language and open source purists express. I care about stability. I care about uptime. I care about speed. But most of all, I care about security. I care about the total cost of the systems I run.

Ok, ok, I'm leaving, I'm getting off your lawn.

Clubcompy.com has an easily accessible interpreter for all as a main goal. I'd add that it's BASIC-like language and virtualized graphics engine makes it pretty easy to write fun games.

(Disclaimer: I'm one of the creators of clubcompy.com.)

Yeah, it's a really good name!

It's probably Stockholm Syndrome, but I'm ... I'm actually feeling sad about this! I spent a ton of time on my site hacking in IE6 support. Just last month I got my compy characters to FINALLY layout correctly in all cases on IE6. Ok, I can't resist a little war story ... In the past, the right hand column of character DIV's had a vertical offset of like 5 pixels. Why? WHY DID IT LAYOUT LIKE THAT?! There's no reason, no known peekaboo bug or whatever that I could figure was the cause ... it was just IE6 getting its digs in. It's like it had planned bugs that only I would see.

Memory un-management, DOM-splosions, layout goofs, CSS head scratchers - it was like trying to carry water with a bucket that has a bunch of rebel army bullet holes in it. One thing I could always count on, IE6's JavaScript implementation was juuust good enough. Me and Resig always had a way to squeak out of the jungle alive.

  IE6: I beat you. I beat you silly countless times. I won! But, I never thought you'd actually die from the beating. It seems you finally have given up the ghost. R.I.P., ancient warrior. As you rot in the 8th circle of hell, I want you to know that while I cursed you and your creators as foul on a daily basis, I secretly enjoyed our time together.

Dave

Please support Touch/Motion and tablets in general, but please don't try to unify the Tablet and Desktop UI's. Why not support two completely different presentation modes, a la Metro? (Just do it better and not so much of a bolt on.)

Who can tell if someone like me wrote the "OBVIOUS" retort above, or not? If I HAD written a lovely missive like that to my presumptive friend jd, I also would have posted AC. It's fun to have a Score 0 actually get read enough, be enjoyed, and get modded up. Getting even a Score 1 is a success. Thankfully, there are enough eye-rolling articles such as this on Slashdot that it's possible to actually have an AC success once in awhile.

Aren't there gobs of address blocks reserved for ISP's to talk to one another? Perhaps ISP's could lead the way by converting their back-channel host addresses over to IPv6 and then release those blocks for public site use? Are there really 2 or 3 billion IPv4 internet addresses serving public clients?

Perhaps IPv4 running out of available addresses is the necessity that will push the experts, at least, to convert over to IPv6...

16GB recommended, or else you'll be waiting quite awhile. But, you could build the thing in less RAM than that, right?

I take it that the "route around failures" and other original design features of TCP/IP and the Internet as a whole relied upon trusting others always having good intentions and cooperating. Those designs were necessary at the time and the reason the internet exists today.

Nowadays distrust, firewalls, and coding defensively is the norm (or it should be). In that light, the internet's design seems creaky and vulnerable.

Do you have any thoughts or feelings on how software has changed and seemingly become so treacherous since you first designed TCP/IP? Would you advocate a ground-up redesign of internet transports and protocols starting with TCP/IP?

I have homebrew business ideas that I've been developing and I wanted to own my own servers and learn how to rack and manage them. I could have rented time on a cloud or PHP hosting site or whatever. But I figure that controlling my server infrastructure means controlling my costs. I consider that to be like owning my means of production if you wanna get all marxist about it.

I'm no sysadmin, but I know enough to get around Linux. I'm not doing an awesome job of it, and I have a big meltdown failure once every two years or so. Usually just a harddrive failure that I can recover from, but sometimes it's more serious. My sites haven't earned enough popularity to get sustained intense internet traffic yet; so far, my boxes have done okay with the occasional big burst of traffic for my sites ( https://clubcompy.com/ and http://cardmeeting.com/ ) that I get from Slashdot or some random blog.

I negotiated my costs as a fixed $150/mo for 4U and throttled monthly bandwidth. And I'm not alone, in the colocation facility I rent at, I see a lot of homebrew rigs racked up with google and yahoo-owned servers (obviously not in the same rack and not as well cooled, heh.) I had no idea what I was doing, and the techs at the facility were totally cool and taught me how to rack my boxes and helped hold them up for me while I mounted them to the rails. The server and network hardware that I have probably totals about 4K and I built them up over years. I've still got 2U free for future expansion. I use only mini-ITX form factor mobos because I want to rack them in teensy enclosures so I can max out my rackspace, and those motherboards run cool so they go for years without any failure - heat kills. I buy passively cooled MB's whenever they're available and still meet my requirements. I have found Intel Atom boards to be extremely reliable in 24/7 operation. CPU-wise they stink, and I wish I could go 64-bit with more RAM, but I just need cheep life support for SATA and ethernet at this stage. I've had DIMM's die before motherboards, I don't mind spending extra for the best manufacturing quality there.

If you have a steady, good paying job and you're a developer, you should have a homebrew project that you hope/wish/dream will someday blow up and become your livelihood. No excuses about cost if you have even a couple hundred dollars a month of discretionary funds to burn. If anything, do it for fun and chalk the costs up to hobby expenses and do it to learn new things. Make it a long term project - over years - and you can pay for it yourself. You don't need magical silicon valley angel vc startup capital to do very cool things on the internet or in wireless apps.

Dave

ok, I know it's fashionable to pile on mozilla these days, however, in this case, I hope you'll be pleasantly surprised to find that mozilla is working with the w3c and google to standardize the concept. The eventual api they have just started to work on is currently being called the "gamepad" api.

Speed the death of Flash and speed the adoption of HTML5+JS as the goto client-side software development platform for most things. The Joystick API is amongst the greatest things coming out of this effort:

https://wiki.mozilla.org/JoystickAPI

Mozilla is doing good stuff here, buck up ppl!

Sigh ... I wondered when the Java plugin was finally going to poop out on us design-wise. Darn thing is so creaky and stuck in web 1.0-mode. Maybe it's time to port CardMeeting to HTML5...

This is a great point. Opera doesn't have a public bugtracker. I didn't know my website was having trouble on Opera until I saw it mentioned in the release notes on Opera 11.10.

CORE-35515 (Opera freezes while loading elements on https://clubcompy.com/ )

Maybe there was something I could do to test and workaround my site freezing Opera, had I known there was an issue at all...

Mozilla's running a public bugtracker is commendable, heck you don't even need to sign up for an account to search for bugs!

But, writing great software is REALLY HARD TO DO! The 10-year old unfixed critical bugs and thousands of unconfirmed bugs is common. Mozilla just wears all that egg on it's face rather than hiding the egg under the trenchcoat like the other guys do.

Ultimately, Firefox will need another complete redesign and rewrite because it collapses under its own weight. Technology and ideas will move past it's current architecture. Sad but true. And then the whole cycle of awesomeness and suckiness can start again.