internet telephony: skype vs. SIP

for a couple of months i’ve been playing around with skype — it works kind of OK. what i don’t like about it is that skype are using proprietary protocols and, thus, one is locked in to whatever support, or lack thereof, you get application-wise from skype…

…which is to say, you are lucky with windows, a bit less lucky with macs, and you get to sweep the floor when you are using linux 🙁 duh! this becomes particularly apparent when you take a look at the sound support: the skype linux client really only support the rather old open sound system (OSS) that has already been flagged as going to disappear from the linux kernel within the next couple of kernel releases. couple that with the fact that it will just take over your sound card without really sharing it with other audio applications and you have a real winner…not.

so, it comes as no surprise that i’ve been looking around to see whether there are better alternatives. there are. i’m currently using sipgate.de, an internet telephony provider operating in germany, the UK, and austria.

the nice thing about sipgate is that they use open protocols and there are lots of internet telephony (or voice-over-IP, VoIP, as it’s also called) applications as well as hardware — meaning, i’ve a number of linux client that i can use (twinkle is the one i like best) but i can also use a little box like the grandstream handytone-486 to hook up my existing analog phone to the wonderful world of VoIP. also, with sipgate i get a free POTS phone number — skype wanted €30 per year for that. now, i’ve got an erlangen phone number and all calls between dorothee and me are just local calls 🙂

sweet.

in bed with flu…

since monday afternoon i’m in bed with flu. not particularly pleasant. fever wouldn’t retreat really until yesterday evening. went to see my very capable physician yesterday who did a very thorough check including a CRP blood test, all in all, revealing that i was probably suffering from a viral infection but on the way back to the living again. he told me to stay at home until friday (which i’m doing).

since monday i had ample opportunity to compare the results of my assortment of thermometers: i’ve accumulated over the years

  • a mercury-in-glass thermometer
  • a digital thermometer in the same form factor as the mercury-in-glass thermometer (and used in the same way)
  • a digital in-ear thermometer (whose head you poke into your ear canal)
  • and yet another digital thermometer, this on to place on your forehead

it turns out that the mercury-in-glass thermometer and the digital thermometer are the most reliable. the other two thermometers produce widely varying readings depending on the angle you poke it into the ear (in ear thermometer) or depending on where exactly on your forehead you place the forehead thermometer — and the trouble is your are unlikely to be able to repeat the next measurement with the same level of accuracy: trying to get two consistent readings in a row with either the forehead or the in-ear thermometer proved next to impossible. both would often report my body temperature as 35.6°C or some other bogus value when both the mercury-in-glass and digital thermometer would tell me that i’d have 38.1°C…

the best one in my opinion is the digital thermometer that replaced the mercury-in-glass ones: it works under the tongue (as does the mercury-in-glass one) but is faster to take the reading: within 90secs you’ve got a reading of the state of the nation (and it beeps a couple of times to alert you that it’s finished taking the reading). the mercury-in-glass has the advantage of being completely analogue and not requiring any batteries, but has the drawback of taking quite a long time (up to 10mins until the temperature had stabilised in some cases).

so, now you know what computer scientists do when they are bed-ridden with the flu…