Clearing out the postbox
Lots of posts circling above Heathrow, none of which I’m going to have time to write, so: a linkdump. Bah. Wish I wasn’t so busy.
- On Call Bald does a short series on VoIP: Part 1 Part 2 Part 3. He and I had a play with SIP last night; the UI for all SIP clients is shite. I’d like to fix it; I thought about hacking on Shtoom but I can’t even get that to do a call. KPhone was successful, but I’m a much better Python hacker, hence Shtoom. Maybe when it can connect properly I’ll start hacking on it.
- ANdy Budd pimps the SonyEricsson K750i as a replacement for the K700i that both he and I own. I think I want one of these.
- Hixie bitches about Ajax and other names like “DHTML“, when you could just say “HTML + script“. I think he’s wrong, partially for marketing reasons which Angry Matt is better at explaining than I, and partially because “Ajax” doesn’t just mean the technologies, but also the mental structure and approach. But this argument has played out all over the place already.
- Glom looks pretty cool.
- Jonathan Riddell’s KDE4 wishlist. I’ve said before (on LugRadio) that I think that Jonathan seems to see the same issues with KDE that I do, and he’s trying to fix them. That’s really good.
- If you have a physical copy of a program you can modify it, legally according to a recent court decision (in the US). Good news.
> Hixie bitches about Ajax
Yesterday I attempted to explain to a marketoid what “Ajax” was. His first reaction was “What, the cleaning stuff?“.
Americans need to research this crap before they rename the universe.
3 hours later
Yeah, I know. That was a bit of a flaw, that one. Then again, no-one complains at the Dutch for naming a football team after a cleaning product. Pronounce it “aye-ax” rather than “ai-jax“, perhaps.
3 hours later
> Maybe when it can connect properly I’ll start hacking on it.
I’d leave user VOIP for a few years. The number of people certain companies have working on large installs is astounding. The amount of money thrown at this is silly, and some of the features they’re using to get an edge are somewhat esoteric (I developed some over the summer for a BigCo).
User VOIP that is reliable is a good while away simply because it’s not available commercially properly yet.
9 hours later
Gary: I’m not sure I agree there. Both On Call Bald and I had signed up to a SIP provider; he had forwarded ports on his firewall but I had not. Once I worked out which things to type into kphone’s preferences window, it seemed to work fine. All the problems I had with kphone could be solved by UI, including having the software sign you up with a SIP provider; shtoom I just couldn’t get to work, but if I could then I suspect that the problems are equally the same. If skype and MSN Messenger and AIM and iCHat can work for everyone, SIP can too, if people delivering softphones start thinking about it in the right way.
18 hours later
I’m not saying it can’t be done, it just can’t be done reliably (at this moment). While it can work most of the time for most people, it’s a million miles away from a point where it could conceivably replace traditional public telephone networks. Even for people with ideal set-ups, problems still occur.
3 days later
Gary: maybe. The idea of transferring data across the internet in time-critical fashion either (a) needs to be solved for movies and shit as well, in which case VoIP reaps the benefit of that solving, or (b) already has been solved for movies and shit (RealPlayer, etc) at which point I see no reason why VoIP can’t work as well…?
3 days later
Well, yes, it would benefit from that being solved. I don’t think real-time network solutions really exist for movies or even the accompanying audio yet; the quality tends to be too low (most will give good audio, or good video, not both), and network congestion is still a problem that I’ve seen every time I’ve tried to watch video over a few minutes in length. But yes, if we can solve that problem, then VoIP will reap the benefits.
3 days later
Gary: there’s the point, though, that “most will give good audio, or good video, not both“. All we need is good audio for VoIP, and if stuff can already do good audio then we’ve solved the problem already, no?
3 days later
I guess so, but even then it’s not at a stage where it is reliable. I think I’ll try to produce a post on VoIP in the next few days that coagulate my thoughts a bit clearer.
4 days later
Gary: I shall look forward to that. My concern is really that the UI for all these apps is completely ridiculous and convoluted and asks loads of stupid questions. The feasibility of the underlying technology I have sort of taken as read…
4 days later
Ah. The UI is something I’ve barely looked at, particularly for computer based apps. When I was working for BigCo, one of my projects meant I had two of their IP phones on my desk (as well as a real phone) so the UI was the same as any other phone (you dial a number). Well, that’s not quite true; they also had natty touch-screen extensions for all sorts of little extras that IP phones enable.
5 days later