We had to provision 2 new 16 disk storage arrays for network backups, and ended up paying about 3x what it should've cost for the storage. Needless to say our department's budget isn't looking as good as it could.

This is a pet peeve of mine so I'm going to point it out, because they made the mistake in the article and you're making the same mistake in your comment. I'm sure that many Nigerians live in poverty, but those numbers don't exist in a vacuum; the average income of a person given in our currency without any other figures is completely meaningless. The cost of living in Nigeria is obviously also drastically less, otherwise I guess half of the 150 million people who live there are going to die in the next few weeks from starvation.

If I want QoS and I need to prioritize traffic I could do it on my own router if there was no "overselling" (I think you mean oversubscribing). Different types of traffic have different priorities to different customers.

For what it's worth, Sharepoint / Office 2010 offers... some... simultaneous document editing. It's pretty limited and isn't nearly as responsive as Google Docs, but it looks like MS has at least been trying.

I would gladly pay more than $15 extra for that level of support.

My ISP has had a problem with what I suspect is a fibre media converter that is causing high packet loss with packet sizes 1350 to 1500 bytes. My friends and I who live in town all set our MTU manually to about 1300 to avoid the problem, but everyone else in town using this ISP is stuck with websites that time out randomly for no reason, web pages that fail to load randomly, etc.

I tried to explain to support that they need to run a ping size sweep on their router so they can see the packet loss but they guy seriously fired up a windows command prompt from his support machine and ran ping with the default arguments to my home IP address, said all 5 packets were okay, and that nothing was wrong.

Actually, there are probably a lot of malware authors giddy at the thought of a legitimate malware notification service. There have already have already been large phone campaigns by botnet creators with the phony premise that the callee's computer is infected, with phony instructions to remove the infection (install new malware, obviously). Once there actually IS a legitimate service doing this it will be even harder for less tech savvy people to tell the difference.

When I was talking about standards based video telephony endpoints I was referring to SIP/H.323 video conferencing suites and hardphones that are video enabled, such as the Cisco 9000 series handsets, and Polycom VSX VC codecs, as well as VC MCUs such as the Polycom RMX series, or the Tandberg (now Cisco acquired) MCUs.Those devices are all hardware devices in which the codecs have been embedded and they are very common in traditional video telephony infrastructures. None of those are going to support VP8.

I can tell you don't have much experience with VoIP or larger enterprise communications / telephony networks, and that's fine, but server infrastructure in a VoIP network provides call authorization control (who can call who, can Bob make long distance calls, can he do it after business hours?), routing (Bob's IP telephone is located on a private network in office A, Carol's IP telephone roams on the Internet, and Judy doesn't have an IP phone and needs to be reached through a PSTN trunk located in office B, etc), PSTN interoperability (bridging calls to conventional POTS networks, which PSTN gateway is used to route a call, many larger businesses may have multiple multi-channel PRIs or standalone POTs lines in different locations with various implications for the DIDs associated with those lines, etc).

I suppose if you're looking at it from the extremely simplified "mom and dad want to talk to each other" standpoint then it's easy to miss things like this, but that wasn't what I was talking about in my original post. End to end proprietary technologies such as Skype and Google Talk already exist for those kinds of users. Businesses generally base their communications on network standard protocols and technology that use infrastructure that they own, for security and control, and in some cases peformance. The word "network" to you might just mean Internet, but many businesses may use a VPN to connect multiple IP telephony that doesn't necessarily run over the Internet.

I did some reading on it and it sounds good but VP8 for video? Seriously? Forget about the billions of dollars invested in existing video telephony infrastructure which made use of standards based codecs like H.261, H.263, H.264 which would now all be incompatible? Doesn't sound like a very promising solution; the only thing it would work with is other WebRTC endpoints, and a ragtag handful of open source VoIP endpoints which also support VP8.

I'm not making the point that you can't write X in javascript; it's a turing complete language, you can obviously write anything in it.

What I'm questioning is whether the performance and bindings are going to be sufficient: You're not going to be able to write an actual video codec in javascript with any reasonable level of performance. You're not going to be able actually send and receive UDP packets from your web page just because you can write an IP stack in javascript, it needs to be supported by the runtime and therefore the browser.

Embracing HTML5 is in some ways like trading one master for another; with Flash we are the mercy of Adobe for security and feature updates, with HTML5 you're at the mercy of the browser. At least Flash is the same in every browser, whereas the streaming protocols / codecs supported for the video tag in HTML5 will vary from browser to browser.

I think people are maybe too quick to predict the demise of Flash.

What I is the demise of flash being used for the wrong things, which is just as good. Flash will no longer be a requirement for video or richer interaction / graphics / animations as HTML5 takes hold, which is a good thing. People are quick to forget in all the HTML5 excitement though there are still plenty of legitimate applications that HTML5 can't do, or at least, won't do very well.

As an example, how about a SIP video softphone accessible from a browser? In Flash you would implement this through an applet that connects to a server application using RTMP (with RTMP over UDP for media) and you have access to a variety of codecs, where the server application performs the actual bridging to SIP destinations and any media transcoding. Is it possible with HTML5? Perhaps, if WebSockets was a mature enough technology and the streaming video / audio codecs were sophisticated enough, but they certainly aren't in the current state of the standard, though I would love to be proven wrong.

What's even better is when your number was either given out as a random fake number by someone with a real debt or you've taken on the number from someone who had real debt. You can tell the collection agency to stop calling and that "Art Vandelay" no longer owns the number but they're all trained to assume you're lying.

Did you know that your hard disk is actually already made out of multiple platters with smaller capacities that make up the whole transparently? Your RAM is made up of dozens of individual smaller chips that make up the total capacity, and so are existing SSDs and USB flash memory sticks.

Kids these days.

Spoken like a true 16 year old.

Just make sure if you buy anything it's cash only. :)