Indeed. KDE does this too, and in fact, when people say that Kopete supports a master password, what they really mean is that it's using the main password store of KDE.
In fact, the inability of Firefox to access the KDE password store is the main reason I now use Konqueror. That, and the way Firefox chooses not to remember some forms, depending on the weather or perhaps the phase of the moon.
OTOH, maybe the addition of VoIP to XMPP will cause more IM clients to support VoIP natively which would make them easier to extend to support SIP - that would be good since there seem to be very few open source soft-phones ATM (I've certainly not found any decent ones, I've been using sjPhone which is closed source but happens to be freely available for Linux)
I dunno, I find Twinkle to be quite good. The only flaw I can pick with it would be its lack of support for video chat.
Okay, Otto... before bitching that Linux can't easily configure and use a Bluetooth camera, how about we point out that neither can Windows? Been there, done that, driver didn't support it.
Moving contacts between servers is a pain, yes, although only if the original server is down. If the original server is still up, then the task is trivially achieved using Jabber Roster Utility which is installed on several public web sites around the place... implementing it in clients would be a nice touch, but nobody wants to.:-)
As for a stable ICQ transport, the new PyICQ-t actually holds up pretty well. It uses a crapload more memory than JIT used to, though.:-/
The point of Jabber isn't the transports anyway. Ideally, people would just stop using the transports so that all those developers can focus on the more important things which would make it really revolutionary. I want PubSub news and weather delivery inside the IM client. The PubSub should be able to deliver video newsfeeds where the video is also delivered through Jabber extensions. Blogging should be possible through IM clients. Etc.
Right. And that's why people said... you can't complain about bugs if you don't report them. If the bug only affects you and you're not a user, then the bug isn't affecting anybody.
Indeed. I'm with a provider that AFAIK doesn't offer emergency call services, and it doesn't really phase me... my ATA actually came with the emergency number preconfigured to route through to the normal landline, which we're effectively forced to buy as it is...
Step 1. Report bug in a clear and consistent fashion in the correct place.
Step 2. Wait for the bug to be fixed.
Step 3. Profit! Well, not really... but you know, it's hard to resist that one.
The calls themselves are most certainly cheaper, though, so I suppose it really depends whether you make a lot of calls, or hardly any calls. If you consistently make interstate calls then there would be a big difference between paying STD rates for every call, vs. paying a tiny flat rate for every call.
I gather that if it were something like Gizmo, for which half these arguments don't even apply, they would simply try to come up with even lamer arguments to use against it.
If you care so much about Qt being proprietary (which it isn't, actually... it's free software), then I'm afraid you'll have to stop making proprietary software yourself, if you expect your arguments to hold any weight. But then again if you weren't making your own software proprietary, you wouldn't have to pay for Qt, would you?
Well, we already have the choice of languages, don't we... it doesn't really matter whether it's done via a universal VM or via a Ruby interpreter, because the result is the same for people who want to use Ruby.:-)
Delete is perfectly reversible. All they would need to do is make a second archive called "Trash". All "deleted" email goes into the trash, you can't empty the trash, but perhaps searching doesn't return hits if the mail is in the trash. Problem solved, and still in a GMail-like fashion.:-)
SKA is in planning, too. For some reason reading this article triggered the "wait, I knew there was a telescope mentioned a while back that had a collecting area of a square kilometre..."
In a way, this would be more stable, wouldn't it? Take those ATI drivers which are perfectly stable on Windows but appalling on Linux. Wrap those in this magic driver wrapper, and wham. You have a driver that runs under another OS. On the other hand, I suppose that running things like graphics drivers through an emulation layer would be ridiculously slow. Wireless networking drivers aren't so bad because there is already a crapload (that's a technical term, btw) of latency already, such that no extra lag can really harm anyone.
I don't see a lot of people griping when their Closed-Source ATI linux driver keeps their video card running on their "open source" OS...
Maybe not many. But I assure you, we do exist.
ATI's shitty closed source driver crashed Xorg long enough for many people to get really pissed off that nobody could fix the problem. The people in the know couldn't fix it because the source wasn't available. The people who made the drivers wouldn't fix it because they assumed nobody was running Xorg. It was a real mess, a horrible piece of work.
I don't know if the mess has stopped, because I stopped using ATI due to it being unworkable. Unfortunately, I'm now on NVIDIA, and finding that running an OpenGL screensaver overnight almost always results in a crashed computer when I check it in the morning.
So, what's the icon for Atom feeds then, if that one's allegedly for RSS ones?
Indeed. KDE does this too, and in fact, when people say that Kopete supports a master password, what they really mean is that it's using the main password store of KDE.
In fact, the inability of Firefox to access the KDE password store is the main reason I now use Konqueror. That, and the way Firefox chooses not to remember some forms, depending on the weather or perhaps the phase of the moon.
OTOH, maybe the addition of VoIP to XMPP will cause more IM clients to support VoIP natively which would make them easier to extend to support SIP - that would be good since there seem to be very few open source soft-phones ATM (I've certainly not found any decent ones, I've been using sjPhone which is closed source but happens to be freely available for Linux)
I dunno, I find Twinkle to be quite good. The only flaw I can pick with it would be its lack of support for video chat.
Obviously the solution to most people's problems is simply not to test the site in IE.
Okay, Otto... before bitching that Linux can't easily configure and use a Bluetooth camera, how about we point out that neither can Windows? Been there, done that, driver didn't support it.
Moving contacts between servers is a pain, yes, although only if the original server is down. If the original server is still up, then the task is trivially achieved using Jabber Roster Utility which is installed on several public web sites around the place... implementing it in clients would be a nice touch, but nobody wants to. :-)
As for a stable ICQ transport, the new PyICQ-t actually holds up pretty well. It uses a crapload more memory than JIT used to, though. :-/
The point of Jabber isn't the transports anyway. Ideally, people would just stop using the transports so that all those developers can focus on the more important things which would make it really revolutionary. I want PubSub news and weather delivery inside the IM client. The PubSub should be able to deliver video newsfeeds where the video is also delivered through Jabber extensions. Blogging should be possible through IM clients. Etc.
Right. And that's why people said... you can't complain about bugs if you don't report them. If the bug only affects you and you're not a user, then the bug isn't affecting anybody.
Indeed. I'm with a provider that AFAIK doesn't offer emergency call services, and it doesn't really phase me... my ATA actually came with the emergency number preconfigured to route through to the normal landline, which we're effectively forced to buy as it is...
Step 1. Report bug in a clear and consistent fashion in the correct place.
Step 2. Wait for the bug to be fixed.
Step 3. Profit! Well, not really... but you know, it's hard to resist that one.
The calls themselves are most certainly cheaper, though, so I suppose it really depends whether you make a lot of calls, or hardly any calls. If you consistently make interstate calls then there would be a big difference between paying STD rates for every call, vs. paying a tiny flat rate for every call.
...they open the source code up.
Thing is, there is an implementation of XUnit in rather a lot of languages, including C and C++.
Use FastCGI instead. There is a general consensus in the Ruby community that mod_ruby should die, and die HARD.
I gather that if it were something like Gizmo, for which half these arguments don't even apply, they would simply try to come up with even lamer arguments to use against it.
That's pretty funny, you just used Microsoft Visual C++ and "quality of tools" in the same sentence. Now everyone knows you're just trolling though.
If you care so much about Qt being proprietary (which it isn't, actually... it's free software), then I'm afraid you'll have to stop making proprietary software yourself, if you expect your arguments to hold any weight. But then again if you weren't making your own software proprietary, you wouldn't have to pay for Qt, would you?
Trolling? Do you seriously think that Microsoft aren't trying to dominate the world?
I guess someone has to use Python, because all the good Python programmers already moved to Ruby.
Well, we already have the choice of languages, don't we... it doesn't really matter whether it's done via a universal VM or via a Ruby interpreter, because the result is the same for people who want to use Ruby. :-)
It would still be better if Ruby were universally supported as a client-side scripting language for web pages.
Crap. Sounds like we need more address space. Anyone for 1024-bit addresses?
Delete is perfectly reversible. All they would need to do is make a second archive called "Trash". All "deleted" email goes into the trash, you can't empty the trash, but perhaps searching doesn't return hits if the mail is in the trash. Problem solved, and still in a GMail-like fashion. :-)
SKA is in planning, too. For some reason reading this article triggered the "wait, I knew there was a telescope mentioned a while back that had a collecting area of a square kilometre..."
In a way, this would be more stable, wouldn't it? Take those ATI drivers which are perfectly stable on Windows but appalling on Linux. Wrap those in this magic driver wrapper, and wham. You have a driver that runs under another OS. On the other hand, I suppose that running things like graphics drivers through an emulation layer would be ridiculously slow. Wireless networking drivers aren't so bad because there is already a crapload (that's a technical term, btw) of latency already, such that no extra lag can really harm anyone.
I don't see a lot of people griping when their Closed-Source ATI linux driver keeps their video card running on their "open source" OS...
Maybe not many. But I assure you, we do exist.
ATI's shitty closed source driver crashed Xorg long enough for many people to get really pissed off that nobody could fix the problem. The people in the know couldn't fix it because the source wasn't available. The people who made the drivers wouldn't fix it because they assumed nobody was running Xorg. It was a real mess, a horrible piece of work.
I don't know if the mess has stopped, because I stopped using ATI due to it being unworkable. Unfortunately, I'm now on NVIDIA, and finding that running an OpenGL screensaver overnight almost always results in a crashed computer when I check it in the morning.
Great stuff, those closed-source drivers.