So I took the disk home and tried it on my older laptop. Success! The wireless stuff worked (not on the live CD though, I think it had to save my WPA password somewhere, but it couldn’t), the wired networking worked, and the graphics worked. These are usually the things that suck. Even the Bluetooth seems to work (not that I have much to test it with). I’m quite impressed. It all just works.

So, much to my surprise (I’d tried various version of Xubuntu and Kubuntu 8.04), I actually quite like it. I guess my impressions of Linux from nearly 10 years ago have become rather out-of-date. I reckon I may actually stick with it. There’s something nice about a decent OS that still has 15Gb free on a 20Gb partition, even with a bunch of stuff installed. If I can set up a nice text editor, I can probably do most of the stuff I need to do, and Wine seems to run some stuff well, although Google Earth sort of killed the computer.

I know it isn’t exactly news to most people that Linux can actually be good. I guess I’ve basically just become lazy — Windows XP was familiar, and worked for me. And I still had those memories of tedious tweaking from years ago. In fact, tedious tweaking is one of the things that I hate. I like stuff that just works. I hate options, that’s why I kind of liked my iBook G4 (if only it weren’t so heavy). I think it’s also why I disliked KDE, GNOME seems to provide much less scope for fiddling.

It was an article on PowerTOP that really got me thinking though. Linux has made some pretty decent advances in recent times — it really is pretty damn good. (I guess all the setting up of SliceHost slices I’ve been doing for work, and getting the Pandora probably also helped push me in this direction.)

And today the news that Google have open-sourced Android. Exciting news indeed — I sense there may have to be some sort of combination of these two things…

What a pity we might be moving offices soon :(

They apparently have a policy of reducing customer interaction (especially phone-based interaction). Perhaps if they processed orders properly, that might reduce customer interaction. What they have at the moment seems to be just a “get the payment, then who cares” mentality. I’m sure a business process analyst would probably explode if they had a look at how much inefficiency there is.

I should have known to steer clear of them, we ordered some stuff at work from them, and that was processed in a similarly terrible fashion (including the return of a motherboard and graphics card when both didn’t actually work). But the price was too compelling… Why do I always buy based on price?

I ordered at 11am today. No news until 5:30pm, when they email (not phone) me to tell me my card details aren’t right and could I please correct them. Considering most other websites manage to do this in real-time, this is pretty lame. But what makes it even more lame is that the card details were right — I can’t correct them because they’re not wrong! My billing address is correct. My phone numbers are correct. What the hell am I supposed to do?

This, combined with the previous experience and their terrible website lead me to think that your shopping is best done elsewhere. If you’re in the UK, and you’re thinking of ordering some computer stuff online, don’t order via DABS. They suck.

Update:

Some links for DABS phone numbers:

• More Trouble with Dabs.com (read the comments, they seem useful)
• DABS.com Delivery Problems
• Yahoo Answers

Then about a year later, I was interested again (these things tend to go in cycles for me; also, the prospect of another UK winter made me want to have something nice to listen to when it’s freezing outside). I found a forum post from a guy in China called Fumac (well, that’s his forum handle anyway). He was building some modules based around the Hypex UcD ideas, but with a much higher switching rate (around 1MHz). It seemed like an interesting idea to me, for various reasons. Also, his modules called MCD, were cheaper and easier to setup (more stuff built-in to the modules, like speaker protection etc, and a power supply). So I ordered some from him.

And they sat around for a few months (along with the transformer I ordered) — suddenly I was very busy at work! Then I bought some speakers (EPOS M12.2 — somewhat similar to these; it seems they’ve been updated, oh well…), and thought I should probably get the amp working. So an entire day of walking between four different Maplins stores got me enough stuff that I could hook something up. And by midnight I had one channel working through a car speaker — terrific!

About a week later (work again!) I had the new speakers hooked up to the modules, and the whole setup sounds wonderful. Actually significantly better (I reckon, non-scientific of course) than the same speakers hooked up to a Roksan Kandy amplifier — this is what I tested in the shop. Bass is terrifically well controlled — subtle, but you can definitely hear the difference between different types of instruments. And, whether it’s the speakers or amplifier, the treble never gets annoying or grating.

I really am pleased with this — the MCD-255 amplifier and speakers sound absolutely magnificent. I’ll post some pictures later…

These two posts on TechCrunch: “Australia joins China in censoring the Internet” and “Australian Government equates freedom of speech to liking kiddie porn”.

Are poorly written, needlessly inflammatory, and in the case of the second one, totally wrong. I think Duncan Riley (the author) seems unable to separate his political leanings from his reporting. TechCrunch isn’t a political blog; it’s meant to be about the web industry. I don’t read it for lame political commentary from people with a primary school approach to politics.

But that’s not the reason I don’t read it anymore. The real reason is that, after writing several comments highlighting my issues with the first article, the second one came along, and posted a comment there. Essentially saying what I’m saying here — perfectly reasonable, perhaps slightly harsh. But considering the state of the second article, and the fact that it totally misinterprets the quotes that form the basis of it, I think being harsh is just being fair.

But my comment disappeared. And was replaced by a comment from Duncan telling one person they were an idiot, and another they were naive (in not those exact words). I would like to give Duncan the benefit of the doubt, especially since part of my comment was talking about the juvenile way the original article was written. But it seems like a strange coincidence to me. Also, I now appear to be blocked.

Lame.

But what else have I been doing? Not blogging, that’s for sure. 250 or so kilometres of that cycling was in France — from Dieppe to Paris, over three days. That was pretty amazing (pretty rainy on occasion too; and certainly hilly) — zooming through quaint French towns that I’d assumed wouldn’t actually exist as they do. 100km was in the Netherlands — from Hoek van Holland to Amsterdam. That was done on one sunny day — pretty much the best cycling I’ve done for a while, actually.

Other stuff? A lot of work — stuff at Moneyspyder is going pretty well, and I’ve learnt a bunch of stuff about Ruby, Rails and e-commerce. I’ve been to a few spots around Britain, Paris (twice); various visitors have come and gone, and I’m halfway to getting a different visa. Summer (as in the weather) in London pretty much sucked. And it’s definitely turned autumn now, so I’m preparing myself for several months of darkness (or at least dullness).

So there you go, a summary of stuff… See you again in six months!

create your own visited countries map

I couldn’t really justify ticking “USA” — I’ve been to Hollywood, and Hawaii. Not the entire rest of the country — and it would highlight just a little too much. I also didn’t tick any places that I’ve only been in the airports of, since that wouldn’t be very sensible. Come to think of it, I haven’t even been to most of the northern half of Australia. Hmmm…

Assuming my total daily commute is about 15km (which it basically is — I go a slightly longer, more scenic way), then that’s 60 days of cycling (12 weeks). If I had caught the bus instead, that would have cost me £120, and taken approximately 30 hours longer (in total, 15 minutes extra each way, on a good day). If I’d caught the tube, it would have cost £240 (and only taken slightly longer on a good day). (Although I guess I would have weekly or monthly passes, which would push the price up, but I wouldn’t have to pay for transport on the weekend). Obviously the cost of the bike has to be taken into account, but I reckon over a year, it will be offset quite a bit. And it’s much more fun.

So cycling is a pretty good option for me — not many downsides, and I get to wander around in the sunlight — and with a much better chance of predicting when I’ll arrive than if I caught the tube or the bus. (Now to get my back wheel fixed — it’s a bit wonky; from potholes I think.)

The weather looked a bit dodgy in the morning, but by 1pm the sun was out fairly frequently, and it got sunnier from there. I tried to use my London Cycle map, but that thing’s pretty big, and requires folding and unfolding (and stopping), so I just did my usual “point in the right direction” style of navigation (plus a few stops to look at the map; more for reassurance than anything else).

I was intending to go down to Bayswater Road, and west from there, then make my way along the bike path next to Holland Park. I didn’t get there, so I sort of ended up on Kensington High Street, and turned south from there. A few random turns later, I was at West Brompton tube; a bit after that I was on the highly trafficed North End Road, where they also seemed to have a market (with the stalls pretty much a meter from passing buses) — so that was pretty slow going. Then I went down past Fulham Broadway, Walham Green, Parsons Green (where the work Christmas dinner was), down New Kings Road, and across Putney Bridge.

From there I headed up to the London Wetlands Centre — the embankment seemed to be taken up by some sort of running event; not quite sure what it was. Then I sort of headed west along a bike path, then followed the random National Cycle Network signs to Richmond Park (it was actually pretty well signposted). (Other than that, I have no idea how I got there…) One of the slower parts of the trip was waiting for two railway level crossings (within 200m of each other). While waiting there for a total of about ten minutes, I chatted to a couple of other cyclists (Richmond Park seems to attract a lot of them!)

When I got to Richmond Park, I stuck mainly to the dirt shared cycle/pedestrian path. It’s good that there’s a park in London that actually lets cyclists in — on more than just the roads. I was pretty happy to be able to wander off into the countryside, although a couple of times I managed to make it onto non-cyclist paths (and I was really trying to look out for the signs!) There were a few people really speeding about, which I thought was a bit ridiculous. (There were also a few mountain bikers wandering off into the countryside; it looked fun, but it’s not allowed. So I guess that’s something I won’t be doing…)

After seeing all the “Warning Deer” signs (but no deer), I was starting to wonder if they actually existed. Well, they did — a couple of herds (or whatever you call a collection of deer) were wandering about, I got a few photos:

After that I helped a guy who’s bike chain was stuffed, and headed out of the park. At the level crossing I ran into the same cyclists I’d seen before; I ended up following them back into the city — they were heading a different way, which I thought might be interesting. So I ended up in Hammersmith; then despite some very good directions, managed to get slightly lost, and end up in Notting Hill. Which didn’t matter too much — at least it’s pretty close to home, and I knew the way.

So all in all, an excellent afternoon out on the bike! (It’s fun to do something other than commuting…)

Probably good for winning votes. Useful? Not particularly. It just seems like a way to introduce inconvenience into people’s lives for no reason. I suppose it could save the odd house-fire (which would be good); it would indeed cut power usage. And there’s the thing; they don’t need to take standby out of electronic goods to make them save electricity. There’s no reason standby needs to be a bad thing.

In fact, each year, companies release microcontrollers that run on less and less power. Those RFID swipe cards you use are a form of microcontroller, and they’re powered by the magnetic field of the reader! Sure, some electronic goods have ridiculously wasteful standby modes; but they needn’t. Your iPod goes into standby when you “turn it off” — Apple don’t even see the need to have actual power off on something that’s battery powered; I’m guessing that the microcontroller is fairly efficient.

The technology is already there; that should be the new law — make companies use current technology; introduce standards on how much power certain goods can use when they’re on standby (and while they’re on for that matter; some things are ridiculously wasteful). Make them ambitious if you like, I’m sure the engineers at electronics companies would like a challenge.

And while you’re at it, introduce requirements to use efficient power supplies (most aren’t); introduce requirements for automatic standby modes on certain equipment; try and reduce the losses in power line transmission (impossible, perhaps, but I don’t really know); and make power stations more efficient!

What’s the point of making a tiny dent in power usage (with the standby thing), when the same gains (and more) could be realised by making our power sources more efficient? This whole thing just seems like a way to seem more green without doing anything difficult. (Actually, telling companies to take stand-by out of things seems pretty difficult to me; but I guess government can just blame corporations when the whole thing fails. In the meantime, they’ve lost the opportunity to actually do something useful, though.)

The insulation thing is a good idea; the lightbulb thing is a good idea; this stand-by thing is just stupid.

You need the Composed Of Conversions plugin. Basically it lets you write a converter block with your composed_of declaration. This means that (to take a wildly relevant example — I’ve been trying to get this to work for a little while now) you can put “AU” in a text box on a form (or a <SELECT>), have that converted to a Country object, then write that to a string column in your database. Seems like a roundabout way of doing things, but it suits what I’m doing at the moment.

Interestingly, the author of that plugin submitted it as a patch to core Rails, but it’s basically been ignored (it seems?). Which is strange, since without it, composed_of is more-or-less useless. (If your attribute has a UI, that is.)

:my_symbol

to something like

---- :my_symbol

Which could be fine. Possibly even useful — hey, I might actually want Symbols stored in the DB like that (assuming I never want to store strings like that…) Unfortunately it doesn’t seem to be paired with Mystery Operation Un-X. In that when it comes back out of the database, nothing is done; you just get a String (with garbage characters chucked in), not a Symbol, and it’s totally useless. My advice: don’t store symbols in the DB.

This could be due to my use of the Composed Of Conversion plugin (which I’m just about to write an entry about), but if so, since it happens one way but not the other, it’s an/yet-another example of deep-embedded Rails code cruftyness.

(Incidentally, if someone can explain this, that would be great. That whole explaining process would probably be made easier if I had comments on this blog… Oh well.)

It’s now my new “I wish I worked here” area — it’s really quite nice. Nice looking pubs and cafes, a park (the grounds of St James Church, I think) — what more could you want? At least it’s nearby, so I can go and hang out there in the sun (when there is sun). Perhaps at night-time it turns into some sort of weird zombie-fest, but given how nice it looks during the day, I’m willing to ignore that possibility.

PS: Sorry I’ve been slack updating this (exciting) blog, lots of work to do… I think I may also split the blog into “tech” and “life” sections — similar to what I’ve seen done elsewhere. I’m reasonably sure that it’s a tiny minority of my readers (which are already a tiny minority in themselves) who’d like to read both, and perhaps I’d put more about living in london if I had a blog/section devoted to it.