jump to navigation

Going offline January 22, 2009

Posted by a1291762 in : Uncategorized , 2comments

In two days time we move house. I will be without internet and my existing cable service will be cut off. While this blog will stay up (it’s hosted on blogspot) everything else will go down. This includes things like images, CSS and JavaScript used on this blog so things might not work so well for a little while.

With any luck, I’ll resume internet service with the same ISP and everything will just start working again (I was told there was a 90 day window for email addresses but I’m not sure about web storage). In any case, I hope to have everything back up in a week or two.

Update 28 Jan 2009

I’ve been with Optus cable for about 4 years now but were they at all helpful in getting me re-connected at my new house? No. Not at all. I’ve felt this before, that they just don’t care about their customers.

Despite telling me they’d keep my account in suspension for 90 days (so I could re-activate it at my new house) they killed my web storage so everything other than this blog is gone.

We’ll be getting Telstra cable sometime next week.

QWERTY vs Dvorak is a lie? January 18, 2009

Posted by a1291762 in : computers , add a comment

Wow. It turns out that in the dim recesses of history, a giant hoax was constructed. Dvorak invented his new keyboard layout then proceeded to try and convert the market. It looks though like the accepted version of history (that luck kept QWERTY in place) is wrong.

There’s 2 parts to this article. The first is that QWERTY is better than Dvorak (or at least, that Dvorak does not have enough of an advantage over QWERTY to convert to it). The second is that the distorted QWERTY myth is kept alive to support anti-market policies.

I can’t comment on the second part of the article (though it doesn’t surprise me in the least to hear it) but I taught myself Dvorak so I can comment on Dvorak vs QWERTY.

I taught myself Dvorak back when I was at Uni. It was painful and slow for about 2 weeks but I think that I eventually met and even surpassed my QWERTY speeds. I say “think” because I never measured my results.

There were 2 major problems with Dvorak.

1) It is not friendly to coding or Unix.
2) Many, many environments have a hard-coded assumption of QWERTY layout. Only a hard-wired Dvorak keyboard can remove QWERTY entirely.

The first comes about due to the focus on typing English words. Some of the important meta-characters are moved about and suddenly writing code (at least, C code) becomes harder. The problem with Unix is that it was designed by lazy people that had QWERTY keyboards. As an example, the simple “ls” command becomes a hand-cramping operation in Dvorak, exactly the kind of thing Dvorak was supposed to avoid! I suspect that if Unix had been created by people using Dvorak keyboards the commands would have different names.

The second problem is a problem because I didn’t have the money to buy a hard-wired Dvorak keyboard and you can’t just plug any old keyboard into any old machine. One of the things I needed to do while learning Dvorak was to banish QWERTY altogether. I replaced the key caps on my keyboard but every time I used any other machine I was again forced to deal with QWERTY. BIOS, Games, etc. tended to avoid the OS-provided keyboard remapping which caused problems.

After I had learned the layout, I memorized the keys and went back to QWERTY key caps. That made swapping between machines a little bit easier but still, most other machines aren’t setup to do key mapping to Dvorak. Strangely, I found it better to cope with Dvorak on my system and QWERTY on other systems than I could with Dvorak and QWERTY on the same system. In the end though, the use of Dvorak on my machine slowed down my QWERTY typing on other machines and I tend to use other machines quite a bit.

So while I did get faster typing Dvorak it was false economy. I wasn’t a particularly good QWERTY typist to start with. I’m much faster now at QWERTY than I was before I learned Dvorak. Faster than I was with Dvorak? Perhaps but I’m not sure. I haven’t used Dvorak in years.

So go and read the article. It’s really long but it’s an interesting deconstruction of a long-believed myth.

Agora bites the dust (thanks to stupid developers) January 17, 2009

Posted by a1291762 in : computers, pre , add a comment

So the Agora (the phone I was most interested in until the Pre was announced) will not be released. Why? Because those stupid developers have made apps that won’t work on its screen.

Let’s back it up for a bit and describe the problem from the beginning.

You can lock the screen size and DPI and let people use pixels. Everyone did this in the older days of computing for speed reasons. Nokia did this with earlier S60 releases (even turning off pixels on newer phones so they’d have the required number of pixels).

You can start small and get bigger. As above but let the desktop scale up. Apps get more room (though they may not use them well) but each increase in pixels must be matched to an increase in physical size. Computers operating systems are still struggling to get rid of this legacy so that we can enjoy high-DPI displays.

You can hide the extra pixels. Palm OS legacy says the screen is 160×160. Newer devices have more pixels but the OS hides this, using the pixels to do things like nicer font rendering. Of course, Palm didn’t completely hide the screen size so you see apps that use the pixels more effectively.

The “correct” approach is to plan to have a variety of screen sizes (both pixels and physical size). Developers must take care to test in a variety of sizes and to handle the sizes properly. The platform needs to provide good support and even (IMO) take some of the burden away from the developer.

An app doesn’t really need to know how many pixels are on the screen yet that’s easy to get from just about any platform. Instead, an app needs to know how big the text is (since the UI should scale with the text), how much physical space is available (ie. how many lines of text fit on the display) and what orientation the screen is in. The layout of the UI might change for landscape/portrait and an optimized version might be used on smaller screens but these things don’t need pixels.

In reality, this is not what I see. In the platforms I’ve coded for you can go to the trouble of determining these things but it’s not automatic and it’s error prone. You need to test multiple resolutions and aspect ratios but that means having a bunch of devices or doing a bunch of serial testing. Why not just give me 4 views of my app when I run it via emulation? 2 portrait screens, 2 landscape screens with big/small and more/fewer pixels.

There’s a system that more closely matches the “correct” behaviour I described above. It’s web pages. Web pages are expected to work with different window sizes and font sizes. The “UI” scales with the fonts. Scrolling of the content is not done by the content but by the frame. What I’ve seen of webOS reminds me of these principles. I wonder if it’ll scale better than what we’ve had before.

Back to my point.

It seems Android developers are doing the wrong thing and making their apps for the G1’s screen. That means the apps will need to be fixed to run properly on the Agora if they can usefully run at all, it all depends on how much support the OS provides. Kogan isn’t exactly a big company. I expect they’ve found out they’re the only ones doing a small-pixel Android phone. I don’t blame them for abandoning the Agora over this, I certainly wouldn’t want to rely on developers to fix their apps for a “little Australian phone” either.

Oh yeah. This doesn’t apply to games. In fact, any app that “renders” content will need to do its own DPI calculations and deal with pixels.

Palm Pre screen is what? January 13, 2009

Posted by a1291762 in : pre , add a comment

About 20 minutes into the massive hands-on video here, Peter Skillman (VP of Product Design at Palm) is commenting on the screen.

“whites are really white, blacks are really white”

Um… I always thought you wanted your blacks to be really black?

;)

We're moving January 13, 2009

Posted by a1291762 in : house , add a comment

We’re selling our house!

The only problem? We now have 13 days to move out and nowhere to go!

It’ll be sad to see the house go. It’s been a great home for the last 4 years. We’ll be moving into a rental for 6-12 months while we build a new house in Eagleby.

Isaac's first day at his new kindy January 13, 2009

Posted by a1291762 in : isaac , add a comment

With Emily starting school at Rivermount College this year, we’ve moved Isaac to the kindy next door. His first day was today and it’s as if he had always been there. He went in with no problems, impressed the staff with his behaviour and was happy when we picked him up.

We’re hoping the next 2 years will let him grow since he won’t have Emily hanging around all the time telling him what to do.

Palm Pre copy and paste? January 13, 2009

Posted by a1291762 in : computers, pre , add a comment

Damn I hate press release journalism. Many articles mention that the Pre supports Copy and Paste. I have seen exactly 1 image showing what appears to be a menu with Copy and Paste items in it. Despite watching every “hands on” video I can find, I have yet to see anyone actually use copy and paste while the camera is rolling.

What’s up with that?

Oh… here it is. Doesn’t look particularly elegant to me however. I mean, trying to highlight text with a finger while holding the shift key? Ew. What happened to one handed operation Palm?

I want a Palm Pre January 11, 2009

Posted by a1291762 in : computers, pre, treo650 , add a comment

So I’ve been happy with my Treo 650 but I’ve also been looking at the new devices coming onto the market. Palm OS is a dead end (I’m pretty sure Palm itself has said this) so migrating onto something else will be on the cards at some point. The question is what to?

The iPhone came out and makes me drool. Unfortunately, it’s expensive and locked down. I would definitely need to jailbreak an iPhone to be happy with it.

In pretty stark contrast to the iPhone’s walled garden is Android. The Kogan Agora is the first Android-running handset that will be released over here (at the end of January) so that’s something I’m keen to look at. Unlike the Horrible G1, the Agora retains a Treo-like form factor. This makes the screen smaller but gives me a keyboard I can type on with one hand.

But wait, there’s another option. The Palm Pre looks like a basically perfect phone to me. The only thing wrong with the launch unit is that we don’t have a CDMA-based network in Australia any more. Luckily there’s a GSM version in the works.

The hardware itself is awesome. The latest of everything, the powerful new OMAP 3. I’ve looked with interest at slide phones in the past but I’ve mostly dismissed them for 2 reasons.

1) The sliders often feel crappy.
2) The top keys on the slider are hard for my fingers to hit (due to the lack of dead space above the top row of keys).

Some sliders have a nice, spring-loaded flick and I’m hoping that’s what the Pre has. For problem number 2, the photos I’ve seen look encouraging but I guess I’ll have to wait and see how good it is.

Of course, the other big problem with sliders is that you can’t use the phone with the slider closed. The Pre doesn’t seem to have this restriction. I don’t use the keyboard on my Treo all the time so I think this will work fine. Of course, as a bonus, you get a phone that is both smaller and has a bigger screen.

From what I’ve seen about the OS it’s going to be much more open, more like Android than iPhone. I especially like that Palm is eating its own dog food here and not giving developers a second-rate environment like Apple did (and still does, via the off-limits private frameworks). The whole “just write a web page” thing really scares me. I have no problems learning a new API but I have tried and failed to make a web page that looked anything like an app. I just hope there’s good templates, standard icons and such. Using JavaScript isn’t really a big deal. It sounds like the platform’s APIs are all exposed to that.

Of course I really hope there’ll be a native API at some point. Palm is talking about a third party providing compatibility for existing apps but that’s only going to happen with native code. Maybe the native code will take the form of a plugin that is accessed via JavaScript?

A big question is how open the OS really is. It seems like it’ll be open from the app side but what about the rest of the system? Will it need a jailbreak? Will it be open source? (I very much doubt it, this “secret sauce” is the only thing that’ll keep Palm ahead of its competition).

I use a Mac at home so you’d think I’d lean towards iPhone. That would be the case if (and only if) an iPhone was as open as a Mac. I’m not an open source zealot so while I’m interested in Android, it’s not for philosophical reasons. Android looks like it’ll get native apps before too long so that platform could be a good place to be. The UI is definitely not up to the benchmark Apple set but I suspect the UI can be improved significantly without a huge effort. Palm Pre easily meets Apple’s benchmark and I think surpasses it by offering features Apple had trouble giving us.

I really love how the Card paradigm works exactly the same as Safari’s tab management in iPhone. It’s so obvious, I can’t believe Apple didn’t think of it already.

To spell it out, I have completely ignored the Blackberry. Too proprietary and locked down for me (apparently that has changed recently so maybe I should give these a look).

The dark horse I see in all of this is Nokia. Since I now work for them I feel like I should at least try to use one of their phones. The only problem? Their phones suck. The low end S40 OS has had features added to it so it’s now as complicated as S60 (Bree hates her new Nokia phone, a very bad omen). S60 itself is slow and complicated. With S60 moving towards open source, with Qt running on S60, with new phones in the pipeline I hope that Nokia can make a device I can like.

In the mean time, I want a Pre. Lets just hope the market doesn’t change between now and the time I can buy one. After all, the Agora is available at the end of January while there is no official date set for a GSM Pre.

We now have TiVo January 11, 2009

Posted by a1291762 in : computers, mythtv, pvr , add a comment

My PVR has died. It freezes up and doesn’t do anything.

Unfortunately, my MythTV box has yet to prove itself capable, let alone easy enough for Bree to use. So since it’s in Australia now, we got a TiVo. They’re being sold as TiVo HD but it seems to be just a Series 3 box.

Based on usage so far, the TiVo is much easier to get around than my old PVR though it actually has less flexibility and is slower to use as a result (eg. things are in menus rather than relying on hardware buttons for various functions).

Surprisingly, the thing I most miss from my old PVR is the bookmark feature. TiVo saves the position of each video when you stop watching but my old box could set arbitrary bookmarks. This and other little niggles reveal that TiVo is designed for a single viewer. I guess that may have been true when it first started out (early adopters were most likely geeks) but these days, I would have thought most people had families and thus multiple viewers. TiVo’s suggestions? Useless when people with different tastes use the same unit (so we’ve turned them off). The now showing list? There’s no way to organise it and with multiple people’s shows, that’s a pain in the arse. There’s KidZone but it’s a separate mode and limits shows by rating rather than by any kind of organisation.

Still, these are relatively minor issues. The addition of recording via EPG makes up for this and the fact that Bree can use it cinches the deal.

It’s not all roses though. My box has reset itself twice now. Once at 2am while I was watching it (the TiVo folks say that’s not supposed to happen) and again last night but I’m not sure when. I suspect it was at 2am again because that’s the time TiVos apply system updates. The TiVo support people think I might have a faulty unit.