Hi Jim.
Why would a Tier One backbone provider (the largest of whom are also last mile service providers now) look any less favorably towards my consuming 1 MB of backbone capacity using Skype than, say, consuming 20 MB while uploading (or downloading) an image to (from) Flickr or YouTube? I should think that they'd be twenty times more delighted with the former than the latter, although who's kiddin' who here? Where would the backbone providers be if all traffic were to cease flowing entirely? (Although that is another discussion in itself, so I won't digress on it here.) But it's not really about this root level of economics, after all now. Is it.
Also, while this makes for an interesting discussion, which I believe we've started having here several times in the past but never quite resolved, I do not subscribe to the argument that either Skype or its end users are getting away with a free ride, since the Internet's backbones are, in the final analysis, funded by subscribers. Are they not? And one would be hard pressed to suggest a method of allocating those subscribers fees once they are paid, which itself is another set of conversations all its own.
You asked: "... what makes Skype "better" than, say, Vonage as a VoIP app?"
Interestingly, I had this same discussion last night. Skype has a tendency to misbehave at times. I'd say this was relatively infrequent, depending on one's tolerance, or brief enough in nature to make it less than deterring (although there was a period of about two or three months last year when I began to wonder).
Sometimes Skype will drop out for an moment or two, or it will begin stuttering or manifest in some weird-sounding echo effect that you'd expect to hear in a 1950s B horror movie, after which the application re-acquires and returns to normal. In the absence of these anomalies, which normally last no longer than a few seconds, however, the sound quality one gets when using Skype is eminently superior to POTS and cellular due to the broader frequency response built into its codecs. It's been characterized as allowing one to experience a sense of immersion, in fact, tantamount to "being there", whereby participants sense they are all collocated.
"Does Vonage ignore standards? Doesn't Skype rely (when terminating at a POTS destination) on standards?
Yes, they both employ wares that conform to lower layer standards, even when not terminating POTS calls. Otherwise, in the case of Skype at least, the Internet's routers would not be able to make any sense of or know where to route the packets being sent. Skype's mystique lies mainly behind encryption and its propriety directory makeup, along with a host of methods it uses through cyphering that have the effect of obfuscating the status of its dedicated servers (of which there are relatively few) and the content it manages, which consists of both voice and routing and administrative (membership list information, time stamping, accounting, etc.) data.
See: file:///D:/Skype/blackhat-eads-skype.pdf
"Malware designed for Windows is targeted at the largest installed base in the world: is that a standard? Maybe not a de jure standard, but certainly a de facto standard."
Absolutely. The cracker community is probably better organized than our ability to defend against them, and they do exchange, and in effect, abide by their own norms. Call them standards, if you wish, although I suspect you'd get an argument from them if you did :)
So those script kiddies who want to make their mark, who want the widest possible success, target Windows: they attack the OS standard. In that sense, their software is absolutely standards-based, because if it weren't, its success (impact) would be diminished."
While I agree there is collaboration amongst malfolk, that shouldn't suggest that any number of uniquely different methods couldn't be employed to hack an OS, so in this respect we differ on the need for a "standard" here, despite the need for respecting the standard that is being attacked.
"Is there a Skype for Windows?"
There is a Skype version tuned uniquely for all popular operating systems, I think.
--
"Getting away from specifics, the principle is that whether you're writing malware or end-user friendly applications, you want to reach the widest possible audience - thus maximizing your chances of success. ... To the extent that an app creates operability or interoperability problems (and thus, functionality problems) by non-adherence to standards, its chances of success are proportionally diminished."
Again, this touches on part of the discussion I had last night. The reason why Skype is chosen far more frequently than Gizmo (another VoIP application with attributes similar to those of Skype, but tailored primarily to WiFi; see: gizmovoip.com ) is NOT because of the standards they do, or do not, abide by. It comes down to the number of members in the group, as a form of validation of Metcalfe's Law. And aside from the fact that it is generally free and easy to use, it is in large part because Skype has hundreds of millions of users (vs. significantly far fewer subscribers using any other VoIP application) that it usually wins, hands down, and not because it employs or ignores standards that are any better or worse than anyone else's.
FAC
------ |
