TaoSpace

My movie setup

Oliver writes about his Torrent Setup for TV Series and asks how others do this. 

The answer (from my side) is: far more primitively. 

I do not have an always-on machine, sitting and downloading stuff. Instead things are downloaded manually to my MacPro. Oliver's use of Dropbox is a master-stroke and one I might just try and copy. It would be the first really practical use for this (pretty neat) service. 

On the MacPro I'm running Connect360, which published all movies to XBox, which is directly connected (by ethernet cable, since the XBox doesn't do wireless-N) to an Airport Extreme basestation. The MacPro connects wirelessly to another Airport Extreme, which is bridged to the first one. The convenience of sitting on the couch with a wireless controller and being able to turn on XBox with it remotely, browse for a movie and play it on my flat screen TV is amazing. 

I am running into issues though. Firstly the XBox is noisy as hell. That's fine when watching a Prison Break marathon, but not so much when watching a latest drama or near-silent movie like No Country for Old Men. Furthermore, having a full-HD TV means you really want to watch HD content, yet this is harder than it seems. Most HD content is encoded using MKV, which the XBox doesn't play. Converting that into h.264 is not straightforward and very time-consuming. You need to demux audio and video, encode each one and then recombine them. If the file is larger than 4Gb is needs to be split first, etc. 

Even after jumping through all these hoops the result is often a h.264 that stutters like crazy (not a function of the network). If Apple upgrades the hardware of the mini, I might buy one. Right now mini's aren't powerful enough to play MKV directly either. 

The solution I'm eying at the moment is Popcorn Hour. This includes torrent downloading and playing of MKVs. It should also be a lot more quiet than the XBox. Now if it could only be made to watch a Dropbox folder...

Automated Mediocrity II

From the people that brought you tacky automatically generated photocollages, now comes crappy automatically generated music.

I dare you to watch the whole of that introduction video!

Seriously though, I am stunned by just how wrong Microsoft keeps getting stuff like this and just how much (obviously cool and difficult) technology it uses to create products that don't seem to seriously target any kind of market at all.

Good point

Mr. Airbag makes a good point. Technology isn't developing along anywhere near the lines it should be:

Look, the two-terrabyte drive isn't supposed to exist yet. We're not supposed to have that kind of technology until I can take a pneumatic tube to work. So either someone has found a way to travel into the future (also part of the list of things you're supposed to be able to do in a two-terrabyte world) or we're all getting gypped.

Shemeckis.

The web just changed

It is hard to over-estimate the importance of this:

Adobe to Publish Flash File Format Specs - ReadWriteWeb

.

Broken

Firstly my apologies for not posting more. Hopefully a more regular schedule will resume shortly.

Just came across this. Kind of funny, but not at all untrue. I've seen posting around the web lately saying that with AJAX and all the toolkits that are available, the web as a platform is now a pleasure to develop for. Well, or stuff to that effect.

This is nonsense.

Just because you can include AJAXy stuff in pages easily, doesn't mean that the whole thing is easy. Just because you can get a basic app to work using Ruby on Rails, doesn't mean that's its all smooth sailing. Reality is all of this stuff is still hard. AJAX still cuts across at least 4 development paradigms: Serverside (whatever platform you pick), Client-side Javascript, Client-side DOM and Client-side CSS.

Website development (i.e. non-web-app stuff), as the tongue-in-cheek graph in the link shows, is still a nightmare. It also isn't much better now than, say, 3 years ago. It might be the best we have, but that doesn't mean it doesn't suck.

CSS' way of looking at the world isn't that of a designer. It's that simple. You can learn to think the way CSS does, and there are many sources out there to help you, but getting for a design to a site just isn't straight-forward at all (to ward off any possible flaming from people: I've been doing CSS design for donkeys years, both commercial and as a hobby).

I'm not saying that it is impossible to do anything with CSS, just that it is way harder than a layout specification language needs to be.

When you are creating web applications, especially the custom ones, used on intranets (of which my company does a lot), it gets way, way, worse. Sure the Javascript toolkits help. A lot even. But functionality that is basic in desktop-development, still take a lot of time to do on the web.
No sliders, no drag and drop, no grid-views, etc. You can simulate all these with the right UI toolkits and all that, but still...if you evaluate the whole stack of technologies objectively it is clear that this stuff really should just "be there" and "just work".

WhatWG, W3C HTML 5 and all that are slow, yes I know. Changing something as large as the web takes time too. Sure.

However, Silverlight and Flex/AIR do provide these things. And a hell of a lot more besides. Flex is almost trivially easy to use, compared to the whole regular web-stack. It is available on all platforms (as is Silverlight) and it does what it needs to do without any fuss.

Will they work for the web at large? Probably not. But you haven't looked at these technologies in depth, you really should now. Check your preconceptions at the door though, and you'll find out just how much better the web could be if we had technologies for it that we actually designed to do the things people nowadays actually want from it...

Thoughtful

It's been a while since I read something as generally thoughtful and provocative as Jeanette Wing's essay on Computational Thinking (PDF link!). It echos a set of ideas I've had for a while, but never really been able to express anywhere near as clearly as she does.

Don't let the title trick you into believing that this is about high-nerdism. You couldn't be further off the mark. She describes an attitude and skill-set that this needed in the world at large. Something to teach children along with reading, writing and arithmetic. This isn't about computers or programming, it is a way of solving problems that appear everywhere in everybodies daily lives.

Very interesting reading.

Pathetic

OK, so Origami is the same as the Ultra Mobile PC.

Big deal.

Microsoft's advertorial web page on it is too pathetic for words. Poorly composed pictures of soccer mom's and turtlenecked hipsters using it outside etc. etc. Intel's site is just as bone headed. Usage scenarios that have been mixed and remixed for years. A big bucket of empty promises of wireless connectivity and seamless synchronization. "Your life in your pocket"-mumbo-jumbo.

How many more times do we have to go through this? It doesn't work. It never did. Unless you have some specifically designed software to run on it, the experience is going to be crap.

Sure it comes with the "Touch pack". Wooow. From videos it is clear to its main function is to launch standard windows programs. Which of course don't fit the small screen and have to display a set of scrollbars. Including a horizontal one! That's just sin. Haven't Intel and Microsoft learned by now that it is devices with a very specific function that are really succesful?

Just look at PDAs. That market has all but disappeared, eventhough more and more features were crammed in to the devices. I see more people running around with PSPs then I see people seriously using PDAs. And I'm not talking kids here either. This is not going to change any time soon. Nokia's Internet Tablet does a lot of the things the UMPC is supposed to do, but does it in a useful form-factor. Sure it doesn't run all your Windows apps, but you have to wonder if people will really use that ability in the UMPC...

Oh and did I mention it weighs a kilo....?

Just give me my eInk-enabled eBook reader with enough battery power to read War and Peace 10 times and I'll go away. Promise!

2006, the year of the browser (again)!

There seems to be a flurry of activity going on around the browser as a platform again.
For the longest time it seems the ideas behind the web and the browser were set in stone, but then things changed dramatically.

First we had the AJAX hype. This is basically about using the current browser concept to the maximum. I've written before that AJAX in itself isn't much of an innovation. It is simply a recognition of what can be done in a browser if you really put your mind to it.

Relic Reborn?

"People should know when they are conquered."

-Quintus in Gladiator

Bjarne Stroustrup, the creator of C++, has just posted a preview of the next version of this ancient behemoth. It will be done in 2009! Hurrah! It should also be easier teach and learn! Hurrah! He even gives an example! Hurrah!

template<class T> using Vec = vector<T,My_alloc<T>>;
Vec<double> v = { 2.3, 1.2, 6.7, 4.5  };
sort(v);
for(auto p = v.begin(); p!=v.end(); ++p)
    cout << *p << endl;

Hurr....Huh?!

Responses like "Too little too late", "Who gives a shit?" and "Where is my Ruby?" are entirely justified in my book.

Google's plans

The web has been rife with speculation of a supposed deal between Sun and Google.
They seem to have a deal on StarOffice, or if you believe most of the commentators deal on Google offering StarOffice as a service. The speculation continues, eventhough it has been denied by Google already. The deal that was actually announced (Googlebar being bundled with Java) is supposed to just be window-dressing, or a taster of things to come. Sure.

The truth is that StarOffice/OpenOffice is a large and relatively monolithic piece of software, that is not be adaptable for use "on the web" (whatever that is supposed to mean!). There will be no browsing to "office.google.com" and being presented with StarOffice in your browser. Technically it is unfeasible and next to this, what would be in it for Google? Also, Google has always been about simplicity and StarOffice isn't simple. Far from it.

What then? Well, what I find way more likely is that Google is building a document storage/management/sharing facility, similar to GMail. Call it GStore. Store up to 2Gb of documents for free, get sharing, review and distribution options for free, all using common web-based protocols. Either a web-upload or WebDAV will do. Google will insert its contextual ads in the UI and Sun will add a "Save to GStore" and "Open from GStore" buttons to StarOffice. Google will "work with Sun" to add the OpenDocument format that StarOffice and OOo use to its list of supported document formats.

And presto. The combo will corner the market for low volume document management and kill off things like YouSendIt in the process. Let's have it!