I have added a new Asterisk blog to Waitara Networks, which can be found here. First up in a few days will be a quick tutorial in regards to the age old problem of distributed Message Waiting Indicator (MWI), and how to tackle it in a distributed Asterisk solution.

There’s been a lot of talk since the release of Windows Vista at the start of 2007, and just how “bad” it is. Many of the people spreading “the word” of just how good it supposedly wasn’t were found out to have never tried it – even to the point where Microsoft performed a market study, and tested a group of people who “didn’t like” Vista, with a “new, upcoming” version of Windows. The study showed that almost all of them loved this new version.

It was actually Vista they were using – proving that Vista’s problems were largely perception, rather than based in any technical understanding. I’ve never struck any major issues in running Vista.

While Vista does – (as does any operating system) – have it’s problems, they certainly are not significant to almost all users. The final release of Vista to market was, admittedly, somewhat rushed, and many of the features initially slated for inclusion with Vista were dropped months before the launch. Some of the things that made it in, lacked some “polish”.

Enter Windows 7 – (oddly enough, it’s official version is 6.1, but that’s a whole different story) – which is proving to be what Vista should have been. Vista with the polish added. I’ve been participating in the Beta program, and running a number of different pre-release versions for some months. I’ve yet to strike a problem while using it.

It’s smooth. It’s fast. And it’s released in October. Check out the Microsoft site above for more information, but from my perspective, Windows 7 is a very special version of Windows – quite possibly the best ever.

Generally speaking, I’m a Linux user – but Microsoft has done a great deal to drastically improve their flagship product. And it wasn’t that bad to begin with.

With the federal government recently moving to commence the tender process for the first stages of the National Broadband Network (NBN), I am gladdened to see that they are taking the issue of backhaul very seriously. Backhaul has been one of the biggest problems with internet speeds in Australia for many years – with or without the NBN.

It is also pleasing to see that the major news outlets are taking the time to consider the costs and implications of what the NBN will bring to the table.

As the above article suggests, for Australia to truly benefit from the NBN, it needs to be universally accepted and used. Several of the major telcos in this country, most notably Optus, have been calling for the structural separation of Telstra into retail and wholesale arms. Up until now, with their monopoly on fixed-line copper in Australia, Telstra has been able to dictate wholesale pricing to all carriers seeking access for data or voice services.

The copper in Australia is old. In some areas it is so bad in condition, or so lacking in quantity, that many people cannot receive xDSL services of any kind on their existing phone lines. Imagine how much it would cost and how long it would take to replace it? But why bother? It’s OLD technology.

The structural separation of Telstra holds many advantages – not the least of which is genuine price competition. However, if the government takes a bold step and mandates a cut-off date for the entire copper network – once the NBN is complete – Telstra will no longer be in the position to price-gouge with their wholesale pricing – they won’t have a product to do it with.

And then every one who wants a slice of the NBN pie will be on a level playing field. That’s a good thing. I certainly hope it happens.

As systems engineers, administrators, and integrators, one of the biggest challenges is collaboration – making it as easy as possible for your users to work together from their desks, wherever they may be. Some people like to work from home, while others are in the office – how do you maintain the user experience across the corporate boundaries?

Further, the complexity increases by orders of magnitude when you want to start collaborating with business partners OUTSIDE of your organisation. Suppliers, customers, vendors. The security and connectivity issues alone are almost mind-numblingly difficult to resolve in many instances.

Enter Google, and their upcoming Google Wave product. The following is a recent presentation made to developers about this astounding solution. It’s about 80 minutes long, but worth the watch – you’ll be amazed at how interoperable the solution is.

One of the biggest criticisms of other Google products, such as Google Docs, is that Google is ultimately holding your data. Wave is truly next generation – or “next wave” if you like – and allows the stunning power of Google Wave to live in your own data centre, rather than Google’s.

The platform will be completely open-source, and will allow organisations to run their own Wave servers, and inter-operate – or “federate” – with other Wave servers from other organisations.

In a couple of words – this product blows me away. Watch the video and see what you think about it.

There’s been a great deal of debate in recent weeks about the proposed $43B(AUD) National Broadband Network, particularly in regards to “naysayers” who wonder how “we’ll ever recoup that amount of money”.

This $43B spend is, quite simply, not about recouping the investment. This is INFRASTRUCTURE. Do we build $500M public freeways and expect that money to be recouped? No, because it’s infrastructure. The point of infrastructure is to provide stimulus for the rest of the economy to become more viable. The $43B won’t be recouped. It doesn’t have to be.

Over the life of this infrastructure, the potential benefit for the rest of the economy is worth well in excess of the $43B initial outlay.

Once the network is built, it’s operation and maintenance will switch to the so-called “NBNCo”, which will need to operate as a profitable enterprise, as it will be privately owned. The government – whose IT policies I have been a critic of – has actually got this right. They are “gifting” this network to Australia, to allow Australia to remain relevant to the world economy in the years ahead.

We currently lag well behind so many countries in broadband speed and penetration. This will bring us right up to the forefront, and give this country the best possible chance to survive in the modern, internet-connected world.

There has been a lot said in recent weeks in regards to the forthcoming National Broadband Network (NBN), and what it will mean for Australia’s future. While it is still very difficult to say exactly what form the network will take, we do know that 90% of all Australian’s will be able to get a 100Mbps connection via a fibre network connection, directly into their premises, with the remaining 10% to get 12Mbps via either a fibre-to-the-node/DSL-based connection, or a satellite dish.

The “typical” Australian broadband user currently has a 1.5Mbps/256kbps ADSL connection, with most paying around $50.00 for the privilege.

Some projections have the new 100Mbps connections typically costing around $100.00 per month – and for some reason, many people are up-in-arms about it.

Sure – that might be double what most people are currently paying, but lets look a little deeper. Aside from people with Naked ADSL2/2+ connections – (which are about 50% higher in cost than “non-naked” services) – people currently paying $50.00 a month for ADSL1 at 1.5Mbps/256kbps, also pay for a local fixed line to carry that service.

Add the costs together. I currently pay $49.90 per month for ADSL1, plus (depending on what calls I make) around $65.00 a month for the fixed line to put that ADSL1 service onto.

That’s $115.00. The NBN fibre network will completely remove any real need I will ever have for a fixed line. I save $15.00 a month, and get 100 times faster internet.

Not so bad. If I really want a fixed-phone, I’ll get a VoIP service – but most of my personal voice communications are currently done via mobile phone – so I might not even bother.

The really smart ISPs will be gearing up with VoIP offerings in preparation for the new network – many already have. VoIP accounts are trivial to set up on something like an Asterisk server, so if you got your VoIP from the same provider who brings you your fibre (which is the logical choice), they’d likely not charge you to have the service – there would be a small fee to have your phone number pointing to their service, and then call carriage costs. That’s it.

So – in summary – faster internet, lower costs, cheaper phone calls. Where’s the problem? Oh yes – Telstra have realised that in 10 years nobody will need their monopolistic copper line system anymore.

For the home user – the cost advantage, combined with the speed increase will be valuable. For business, particularly small business, the ability to compete with the big boys with speeds they previously could only dream of, and with access to advanced telephony they could never contemplate before now, it will drive the entire economy to new scales of efficiency.

Big business will need to sharpen their pencils and provide REAL service – because the little guys who already focus on good old fashioned service, will suddenly become more accessible to everyone. Globally.

That’s got to be a good thing.

Yesterday’s announcement of the scrapping of the Request for Proposal (RFP) process for Australia’s upcoming National Broadband Network (NBN), and therefore the federal government’s plans to go it alone in building the network has raised a lot of questions, and sparked much heated discussion within the industry.

Without going into the “good versus bad” aspects of the plan, I thought it timely to elaborate on what a properly designed, constructed, and (most importantly) legislated NBN means. What’s in it for home and business users?

Starting with business users first of all – you’ll be able to get a fixed, hard-wired (via optical fibre) connection of up to 100Mbps. Fast? Certainly – it’s as fast as the local network in most small to medium businesses today. The big question will be contention ratios – will the upstream backhaul networks be able to cope with the massive increase in data.

I’m sure that that network will be designed with this in mind – so I think we are safe there. However, international backhaul will be equally important. Australia is somewhat lacking in international “firepower” in this regard at the moment, so this will need to be considered also. Certainly, PIPE Networks new submarine cable PPC-1, currently under construction to Guam – (and directly connecting to the US backbone) – will become very necessary. More such cables may be required as the build out of the network continues.

The biggest advantage in laying out the NBN won’t just be related to the carriage of internet and other data. Advanced telephony services (such as VoIP) will finally have the raw bandwidth required to make them a REAL challenger to existing fixed line copper-based services. The advantages of such services were lost to small, and even to medium sized businesses, due to the bandwidth requirements adversely affecting call quality. This problem will practically disappear.

For the home user – costs should be able to come down. Instead of paying $50.00 every month for an internet service at a fraction of the proposed speeds – and bundled with web space, email accounts, and all the other rubbish – people will be able to choose what they want.

Just want data, don’t need web space, and using a Gmail account for email? Just pay for data – you’ll probably get 12Mbps – the previous minimum requirement for the NBN tender – for around $25.00 a month once connected. Nice.

If you want those features from your ISP – you can pay for them in addition. A-la-carte – exactly what you want, and nothing you don’t need. This helps the ISPs with economies of scale for their add-on services also.

Power users – who might run their own mail and web servers – can finally get cheap, serious bandwidth, and not have to worry about all the frills that are normally tacked on for home users.

The end result will be a very competitive SERVICES market – completely free of worrying about data carriage. Such competition can only be good for innovation and allowing market forces to determine the right price.

Steve Gibson, of Gibson Research Corporation spoke recently to Leo Laporte on their regular “Security Now!” podcast in regards to a trojan that has been doing the rounds of the internet for quite some time.

This trojan – and others like it – infect networked computers via any number of known security exploits, and change the DNS servers that the infected machine looks up on the internet for the purposes of name resolution. The DNS servers the infected machines look up are set up to maliciously redirect you to websites, other than the one you are really looking to visit. They might look like the site you wanted – (eg: your banking website) – but really, they are hosted by the bad guys, seeking to steal your information.

There are a number of ways people – and ISPs – can alleviate this problem on their networks. Generally, this is a problem that should be addressed by ISPs.

It is as simple as blocking all outgoing DNS requests. This forces the client machines to use the local DNS server – (ie: the ISPs DNS server) – which should then be heavily fortified and rigourously secured by administrators at the ISP. They can then configure forward lookups outside of their network to DNS servers that they themselves trust.

That way, when a bad guy’s trojan comes along and configures your client machine to look up their “bad” DNS server, your ISP blocks your access to it by default, so you don’t get their “bad” information.

Any administrator worth their salt would have firewalling in both the inbound AND outbound directions to deal with this – and similar – kinds of attack.

It’s a fairly simple solution to a serious problem.

Sender Policy Framework (SPF) is a system whereby special records – of the TXT type – are placed into the Domain Name System (DNS) to specify the IP addresses of all mail servers that are allowed to send email to the public internet, for that particular domain.

Receiving mail servers can look at an incoming email – from someone@domain.com for example – and then look up the SPF records for domain.com. If the IP address from which the email actually arrived is not in the list of authorised servers from the SPF records the receiving server looks at, it can be assumed that the email is not really from someone from within that domain.

Unfortunately, you cannot say that if an email fails an SPF check that it is DEFINITELY spam. It’s only an indication that it is MORE LIKELY to be spam.

Also unfortunate, is that SPF is not as widely adopted as would be useful. If every single domain name was forced to use SPF, and the administrators of each domain rigidly enforced policy to ensure that email is only sent via their authorised servers, SPF will stop spam dead. Completely.

So, is SPF worth implementing? For the most part, the answer is yes. If you want to give other servers on the internet the opportunity to verify that a particular email is from your domain, it’s the best currently available option.

The biggest problem is, not all RECEIVING email servers choose to do SPF checking on incoming mail. Because it’s not compulsory, SPF lacks the punch it would otherwise have.

Basically, everyone SHOULD use SPF. In particular, if your site is an e-commerce site, it would be responsible to add SPF records for your domain. Because e-commerce sites are more likely to fall prey to phishing attacks, it would be extremely useful to your users to offer this option to receiving servers, to help alleviate the world of phishing scams.

It is up to administrators everywhere to look at using SPF lookups on their mail servers to help end users identify spam. SPF is another string to the bow.

Various news and internet security agencies are reporting between 35% and 75% of internet email spam has dropped over the last few days, after a rogue San Francisco/Bay Area ISP, McColo, was disconnected from the internet by its upstream provider.

After reports of large amounts of suspicious traffic coming from its data centre, McColo had its internet access provisions terminated, resulting in the almost instantaneous drop in global spam traffic.

On a personal level, I’ve noticed that the total amount of email traffic hitting my own mail server has dropped to almost zero. Normally about 95% of my SMTP traffic is spam.

Unfortunately, it is likely that the drop will only be temporary – the spammers will not take long to set themselves up elsewhere – but this victory for anti-spam crusaders demonstrates that a global approach to the problem can have a significant effect on the problem.