New iPod battery

▼ This week, I installed a new battery in my fifth generation iPod. I don't remember when exactly I bought that iPod, but it was probably in spring or summer 2006. So this iPod is more than five years old. When I tested the old battery, I was able to play music continuously for 3 hours and 13 minutes. So that's not even that terrible. However, the trouble was that spinning up the harddisk (yes, the insides of the iPod are almost completely filled with a 1.8" HDD) was becoming a big issue. Unless the battery was pretty much fully charged, spinning up the drive when the iPod booted wouldn't work. Even with USB power connected. (It did work with the iPod connected to a Firewire port, which provides much more power.)

So when connecting the iPod to my computer, it would go through a boot/spin/fail/wait cycle for more than an hour. After that hour it would still sync and be usable as an external drive. Even though I use my iPhone for music and podcasts these days, sometimes it's useful to have a second music player. And playing back podcasts is actually better in some ways on the 5G iPod than on the iPhone, as it simply stops playing after a podcast episode, and the clickwheel allows for easy scrubbing through an episode.

So when I discovered that I could get a new battery for about 15 euros (with shipping) from Amazon UK, I didn't hesitate. (They're available from lots of places online.) The battery even came with a plastic tool for opening up the iPod. It took me a few minutes to connect the new battery and close up the iPod, but all in all the procedure wasn't too difficult.

After the replacement, I cued up a bunch of podcasts, and listened to those over the course of 4 days. This is a much more demanding use case than playing music, as the iPod can buffer several songs in memory and only spin up its drive every 25 minutes or so when playing music. When playing long podcasts, it has to spin up more often, and also when resuming playback later. After some time the iPod goes into some kind of deep sleep so it has to boot up again when you want to use it the next day. I was able to play 6 hours and 53 minutes worth of podcasts after having charged the battery for the first time. That's a reasonable amount of time. But the good thing is that the iPod will now behave normally when connecting it to the computer, even if the battery is completely drained.

So now my biggest worry is that in 2017 when I need to replace this battery, they won't be available anymore...

Permalink - posted 2012-03-11

The mute switch issue

▼

The past week there's been a lot of blog talk about an iPhone going off during a performance of the New York Philharmonic. Apparently someone had just had his Blackberry replaced with an iPhone, and when the iPhone alarm went off despite the "ringer/silent" switch set to silent, the poor new iPhone user first of all didn't notice it was his phone interrupting the performance at first, and then had a hard time killing the alarm.

Cue a series of blog posts arguing for and against the situation where alarms go off even if the ringer/silent switch is set to silent. Andy Ihnatko is strongly against with John Gruber, and Marco Arment to a slightly lesser degree, in favor.

With the ringer/silent switch (let's just call it the mute switch) set to silent, the iPhone will play audio and video with sound, and third party developers are also allowed to play sounds if this is the direct result of the user's intent. All other sounds, with one exception, are silenced. That exception is alarms and timers set through the clock utility. (Alarms set in the calendar application are silenced.)

The reason for this is obvious to anyone who uses the mute switch in practice: it's way too easy to forget to turn the ringer back on. Oversleeping because your alarm didn't go off as a result would be a very bad thing. And this brings us to the reason why what happened at the New York Philharmonic is so unusual: as a rule, people don't set alarms to go off in the middle of a performance they plan to attend. So having alarms sound even if the phone is set to be silent shouldn't lead to trouble in most cases.

Some people want Apple to add settings to customize this behavior. As a rule, Apple doesn't like to add settings, although you wouldn't know it from the iPhone settings tool. Looking at the notification settings, which are quite unwieldy, it's obvious to me that having something like that for the mute switch behavior for different kinds of sounds wouldn't be very useful. Even for people who are prepared to go in and change the defaults, it's just too much work to get everything configured just right. Especially when the consequences of mistakes can be significant.

But I have a different suggestion. When the user sets the switch to "silent", a requester should be shown that says "an alarm will sound at 7:00" (if you've set your alarm to go off at 7, of course). With that, the ringer volume should be shown. This allows users to turn down the ringer volume to zero. (Unfortunately, as things are now, it's not possible to turn the ringer volume down all the way to silent, although at the minimum volume it's not going to interrupt a live performance.)

I think this will give users the info they need to correct possible mistakes without making life for average users more complex. On his podcast, John Gruber argued against a solution like this because users may turn the switch without looking. It's true that this not a 100% solution, mistakes are still possible. But if users know that when flipping the mute switch a warning about alarms will show it's easy enough to glance at the screen if they're even a little bit concerned about alarms. And after doing this the first few times while looking, they know that alarms will sound even when the switch is set to "silent".

That said, I wouldn't mind having a feature on my iPhone that I had years ago on my Nokia phone, where you could set a profile (sounds yes/no, vibration yes/no) and have it switch back to the default profile at a a given time. Really useful if you're going to the movies or some such and need the phone to be quiet for some fixed amount of time.

Permalink - posted 2012-01-19

Overheid begrijpt certificaatprobleem niet

▼ Minister Donner ondernam in de nacht van vrijdag op zaterdag de zeer ongebruikelijke stap om na middernacht een persconferentie te houden. (Zie Overheid zegt vertrouwen in de certificaten van Diginotar op.) Het ging dan ook om een urgente kwestie: er zijn mogelijk frauduleuze certificaten op naam van overheidswebsites uitgegeven. Daarmee kan iemand zo'n overheidswebsite perfect nabouwen en mensen wachtwoorden en andere gevoelige informatie laten intypen.

Maar vreemd genoeg waarschuwt de overheid alleen om niet verder te gaan met een transactie als er een waarschuwing komt. En die komt lang niet altijd. Naar aanleiding van de problemen bij Diginotar hebben Microsoft (Internet Explorer), Mozilla (Firefox) en Google (Chrome) updates uitgebracht die de Diginotar en Staat der Nederlanden certificate authorities niet meer vertrouwen. Dat betekent dat alle certificaten die door Diginotar op eigen naam en die van de overheid uitgegeven zijn, inclusief de frauduleuze, niet meer vertrouwd worden. Mensen die die websites bezoeken krijgen dus een waarschuwing als ze, al dan niet automatisch, de update geïnstalleerd hebben.

Maar degenen die automatische updates onder Windows XP of voor Firefox uit hebben staan, een oudere browser gebruiken, of Safari van Apple gebruiken, vertrouwen de certifcaten nog steeds en zullen dus geen waarschuwing krijgen. Vreemd genoeg negeert de overheid deze groep, die dus wanneer frauduleuze certificaten ingezet zouden worden (en er moeten dan nog meer dingen aan de hand zijn, zoals het onderscheppen van domeinnamen) wachtwoorden en andere gevoelige gegevens afgetroggeld zou kunnen worden.

Wat de overheid had moeten doen was alle overheidswebsites die met het getroffen type certificaten beveiligd zijn tijdelijk afsluiten totdat nieuwe certificaten ingezet kunnen worden, en hopelijk ook Apple een update uitgebracht heeft. Ik beveel iedereen aan om de komende weken niet in te loggen op overheidswebsites, en als je het doet, dan alleen thuis of op het werk, dus niet mobiel of via een hotspot. Het feit dat je geen waarschuwing krijgt zegt helemaal niets.

Permalink - posted 2011-09-04