Telstra's Letter to Shareholders – a lot of talk, no real explaination

Last week, Telstra got booted from the National Broadband Network process, where the Australian Federal Government will be spending about half the amount of money they (wasted) on the bogan bonus to fund/subsidise an improvement in Australia’s broadband capacity down to the last mile.

Telstra, the largest telco in the country, apparently didn’t like the concept that some Govt money in the process might mean the new network actually involves competition between the network, wholesale and retail divisions of the value chain: one network with cost recovery, then a number of wholesalers, and then lots of retailers. You know, competition on a utility.

Anyway, they submitted an incomplete report, and the Govt kicked them out of the process for non-compliance. Their share price then copped a hiding, double digit losses even in a rising market, since it is pretty clear this new network is going to be “where its at” for the next generation of Australian fixed-line internet access. The government will need to legislate to ensure the successful builder gets access to the copper and other Telstra infrastructure, and it would have been much better for them to have been in the game. But they had a dummy spit, and now the pressure has been on for them to explain their incompetence to their shareholders, including yours truly.

Here’s the letter. A lot of talk, but no explaination. What a disappointment.

Continue reading

The Perils of Credit (cards)

On a recent trip that included China, half a dozen stops in Europe, and some time in New York and San Francisco, someone skimmed my corporate credit card. My bank, Bankwest, has a pretty impressive fraud detection system, and when someone (I think based in China) made a $1 transaction (so as to test whether the number was valid, before either selling the details on or coming back to try and clean me out), the bank’s team caught it and gave me a call.

Unfortunately, I was in the US at the time, and for some reason they didn’t see fit to leave a message, so all of a sudden, my card stopped working, but it was at the tail end of the trip, and I decided I’d look into it when I got back, thinking it might have just been a normal effect of running out of credit limit or something.

After getting home and calling the bank, I was told all about the fraud, that the card had been deactivated, and that they’d send me a new one.

With a number of automatic transactions with suppliers now rejecting, there was only a little bit of hassle involved with telling them the new number. Everyone except TPG internet.

With the new number in hand, I dutifully logged into their web control panel and updated the details. The system accepted them, but a short time later sent me an email and told me that the card wasn’t valid. I logged in again, thinking I must have mistyped a number or two, but again, after a while, an email came through rejecting things.

Not I thought they might have had something wrong on their end, as I’d been buying fuel for the car and paying other vendors successfully with it, and I made a note to come back in a day or two to try again.

This went on for a little while, until I got a call from the bank after I’d made a number of large transactions internationally in a short period of time. While confirming the transactions were legitimate, I happened to ask the caller from the bank whether she could see if there was anything wrong with the card that would explain why TPG’s transactions were failing.

She looked back through the records, only to see that TPG was trying to process my card with an expiry date of 20/22, ie, a month much greater than 12 in the year 2022.

When I got my email reminder from TPG accounts the next day, I replied and explained in some detail what I’d discovered, and asked if someone could give me a call or reply via email with an update on what’s happening at their end.

Later that day, upon getting to my home office, I tried to log on, only to find I’d been disconnected. I called through to their phone number on the bottom of their accounts email, and selected accounts, only to find I’d called 10 mins after they’d closed for the day. Of course, they then hung up on me. So I rang again pressed 2 for tech support, thinking I’d play dumb about my connection not working, and hope they’d take my credit card over the phone and reactivate or unblock the account. Instead of being placed in the support queue, however, I was just told they couldn’t help me right now (presumably too many people in the queue, but the system didn’t say that) and promptly hung up on me. Customer service: FAIL.

Giving up, I called the next morning, and after 15 mins on hold for accounts, I told my story to an outsourced phone support person. I tried to explain my situation with some difficulty because of language issues, and she tried to take the CC payment over the phone. Again, it failed. She put me on hold, and then 10 minutes later, hung up on me. Repeating the process, and now getting having been at it for almost an hour by this time, I am finally asked some details about the card, like the bank it is with. “Oh, we’re sorry, we’re having problems with some Bankwest cards”.

So, while on one hand I’m frustrated by the scammers who nicked my details and made this all necessary, I’m actually more frustrated with TPG. Their system had a problem, and they then disconnected me and made me jump through hoops to try and diagnose and solve it, and even with the accurate information, it still took an hour to have someone recognise it.

They’ve now demanded I give them a different credit card, or access to draw money out of my bank account, but I’ve told them to get stuffed – they should fix their system, and email me when they’ve done it. No one else I’m working with is having any problems, and given the quality of their website – design last touched in 1997 or something I think – and their poor quality customer service, I wouldn’t be surprised if they were massively cutting corners on their payment processing gateway: there’s some things you just shouldn’t scrimp on.

If they do this again, I’m off to Internode, who I heard nothing but great stories about while at OSDC last week.

Trademark Applications in AU and US

I recently got a letter from a law firm in the US, alleging that my new business, Hiive Systems, has infringed their trademark. Unfortunately for them, the best they could do in claiming a trade mark was allege our logo was similar to theirs (we don’t think it is), and even though our businesses are in different industries, they demanded we stop using our name, logo and relinquish our registered domains, or they’ll sue us under some loose Common Law precedent formed in 1870 or so when the US government sued a counterfeiter.

Anyway, this is all going to play out over the next few months, so I won’t name names or go into details here lest I prejudice things or give these guys cause to make the issue (more) personal, but the upshot was that I did one of those important and non urgent things, and registered my trademark for Hiive Systems, initially in the US and Australia, and soon via the Madrid protocol.

For the benefit of those that haven’t gone through the process, but have been thinking about it, here are a couple of notes/observations. The key point is that it isn’t that expensive, and it isn’t that hard – if you’ve got a brand you care about, you really should be protecting your rights through a trade mark.

Australian Trade Marks (via ipaustralia.gov.au)

The process of applying for the trademark was fairly straight forward. The online application form was quite user friendly, and they made it easy to upload files and search for categories you want your trade mark applied to, from within the application process/form.

What surprised me the most was the vast array of detailed classifications for trademark categories: while I had previously thought the applications were for around 45 “high level” classes of registration (such as Class 9 which includes computer software), there are hundreds of sub-classes, and each one you select using a checkbox costs you more ($120 a pop). For example, “software” as a search phrase returned more than 80 sub-classes of Class 9 alone.

Starting small, we’ve spent $240AUD on an application for two categories, payable by credit card, and now we wait for the examination process to unfold. So far, pretty easy, and if/when they get accepted, we’ll be up for $250/class to actually register the marks, and then $300/class/year to maintain them: a total startup cost of $370/class.

IPAustralia is currently estimating less than a 3 month assessment period, which is nice – if I’m lucky I’ll know the result by Valentines Day.

United States Trade Marks (via www.uspto.gov)

The US Trade Mark system was decidedly less user-friendly, however, with patience and reading the fine-print, it worked out well.

Much like the Australian example, there are hundreds of sub-classes that you can apply against with your Trade Mark application, and each of these has a cost attached: US$275 per sub class, using the online TEAS-Plus system, which is a little more demanding of what you give it (and which presumably saves the examiners time). The good news is that these fees also appear to cover the actual registration if the mark is accepted, which is a good thing, since they’re a fair amount higher than the Australian application fees.

One snag I got stuck on while registering was the need to declare the reason for the registration for each class – it was important to select one box at a time, and in my case, choose 1a) since we’re already trading, and then upload some proof that we were already using the mark, in our case, a screenshot from our application. This form and button processes was very poor on user friendliness, and I had to delete and re-add records after accidentally setting them to be 1b) (not in use yet) and start over.

The email from the USPTO states that it is likely to be about 5 months until the application even gets examined, however, when it comes to challenges on validity of the trademark, they’ll use the application date as their reference point. If I’m lucky, I’ll know the result before the start of next year’s ski season.

If YouTube's struggling, what hope's anyone else got?

Was just reading a post from Mark Cuban, Blog Maverick, about what he percieves as desperation on the part of Google when it comes to monetising all that User Generated Content on their site.

Frankly, I don’t have the time to be a big YouTube watcher, so I’ll take Mark’s comments about changes and user experience at face value. Regardless though, I’ve long been a skeptic of the “we’ll build a loss making business with lots of eyeballs, and work out how to make a buck from it later” approach used by so many Web 2.0 consumer web plays.

And if Google can’t seem to make this work, you’ve got to ask yourself: “what hope does anyone else have”?

As remarked by Mike Cannon-Brooks in a SiliconBeachAustralia podcast, the sheer volume of eyeballs required to generate a sustainable business, assuming generous CPM’s, is amazing. If you want to have a $10m revenue business, you need to make close to $1m/month. If you manage to get $10 CPM, then you need 100 million page views per month – this is pretty massive scale, all for a business that could afford to employ, at most 50 staff, and that’s assuming you’re generating all of your own content (no licencing costs), and that your server and bandwidth costs aren’t too high.

Of course, this doesn’t mean is isn’t possible (divide all numbers by 10 if you and a few mates want to make it as a small crew), but unless you’re able to drive up your CPM by having content close to a transaction, in the current economy, its going to be an uphill battle.

1800 and 1300 Numbers in Australia

I’ve been doing a bit of research lately on phone setup for my new startup, Hiive Systems. Part of the establishment work is around phone numbers, and I’ve learned a few things I thought might be worth sharing.

Firstly, I’ve learned that – in Australia anyway – VoIP services haven’t yet come of age. Even after iPrimus came and gave a presentation at the Australian Telecommunication Users Group about their new VoIP services (and how their roll-out for Rebel Sport saved a lot of money and made them much more efficient), the presenter admitted that it is really about point to point type services where you’ve got multiple offices.

Secondly, I’ve learned a bit more about the 1300 and 1800 phone number system. We’ve had a 1800 number with Internetrix for a while, and we got it as part of a long distance plan through AAPT some time ago. In setting things up with Hiive, I wanted to have another look at how it all works.

Turns out, there are lots of companies out there that specialise in these sorts of services. Generally, you pay a monthly fee, usually with no committment, and then you pay a price per minute to receive calls. The prices of two companies, AllTel and Telcoworx are included in the attached spreadsheet, which you can see by clicking here.

These companies will route your call to any landline – or, for a higher cost, mobile – number you like, which makes sense. What surprised me the most, however, is that the difference between 1300 (cheaper, because the caller pays something) and 1800 (free to the caller) is bugger all. So, given 1800 numbers look a bit better/more professional/generous, we’ll be getting one of those.

The other thing that was new info is how the numbers are allocated. Since this is a national namespace, it works a bit like domain names, and some numbers – which can spell your company name on the handset – cost more than others. There’s a Federal body responsible for auctioning off numbers that are more in-demand, usually because they use a lot of repetition – called SmartNumbers which appear to be a part of the Australian Communications and Media Authority.

What surprised me was how they price these numbers. By going to the search form on the SmartNumbers website, you can do a search for the phone number you want, and they’ll tell you if it is a premium number, which they’ve determined will have a reserve price at an auction; the more repeating numbers, the higher the reserve it would appear.

As a result, changing the leading digit at the beginning of the number I’m looking for – I’ll keep it a secret for the moment until I’ve won the auction – changes the price from being a reserve of $500 to, in one case, a reserve of $12,500 – that’s a 25x increase in reserve!

Also like domain names, it would appear that once you’ve got one of these numbers, you become the “Rights of Use” holder; I think this is a bit like a lease on a domain name; you never own the domain, but you have the rights to it exclusively.

I remember a Fourth Estate Domain presentation some time ago with the (younger) Jack Singleton, who runs/ran Phone Names, and was trying to tap into this gold rush aspect. While it looks like the Feds have wised up with an auction system to stop companies “claiming” the rights to thousands of names, just to on-sell them like domain squatters, the more open nature of the process now should make people in business think about reserving a number that matches their business name.

OSDC Sydney 2008 – a wrap

I was lucky enough to spend the better part of last week at the Open Source Developers Conference, or OSDC, held in Sydney between Wed 3rd and Fri 5th of December.

I wasn’t sure what to expect from my first OSDC conference – they’ve been running for five years, and we’ve had at least one rep from Internetrix, and now Hiive Systems, head along since the 2nd one.

Users, not contributors

From a business perspective, I’m a big fan of Open Source Software. As the provider of a range of hosted, software-as-a-service style offerings, however, I’ve been a little constrained in what I’ve been able to have our talented team of developers put back into the Open Source community; so far, it has just been participation and help in forums and such.

While it is a legitimate refrain from members of the Open Source community to ask businesses to Open Source their applications and make them freely available, the service/hosting nature of our business makes this impossible. For example, we’ve already had a Chinese based company knock off our entire website design, and if we shared our code verbatim, we’d likely have knock-offs appearing everywhere and destroying any chance we’ve got to recover our significant, sunken investment. For the clients that licence our software to run on their own servers, however, we provide them with access to the source code; partly its a peace of mind things given some of our clients are very large, and partly its about giving them control of their own destiny. They of course undertake not to redistribute it.

So, given we don’t open-source our apps, I wasn’t sure if I would be burnt at the stake or derided as a leech who takes and doesn’t give anything back when I arrived on Wednesday.

Open in more than code

As it turned out, while the community quite rightly has a level of religious zeal towards everything being free and available, there was a much stronger sense of support between participants in the community. The desire to provide mutual support to others that came through much more strongly than I’d ever seen at a conference, in both the presented sessions by day and social activities of an evening. In this sense, the emphasis was much more on the free and open exchange of ideas and suggestions, and there most sessions I participated had presenters asking for tips and suggestions on how to improve their own approaches to solving problems.

Making our contribution

One of the sessions that surprised me the most was a business session run by MySQL guru, Arjen Lentz from Open Query. Arjen’s previous two scheduled sessions were full of great advice and tips – at both a high and at a detailed level – on how we can build a better business with MySQL. Needless to say I had high expectations coming into his third presentation on the Friday afternoon, which continued the concept of architecting for success, just in case. Arjen said straight up that he didn’t expect to have three talks accepted to the proceedings, and that the session would be much more of a forum style chat about business things. Unfortunately, the content wasn’t very good, and it seemed to highlight somewhere we may be able contribute and give something back: in the business sphere of things. Hopefully, the siliconbeachaustralia.org initiative will help provide a welcoming, friendly environment for excellent technologists to deepen their entrepreneurial and business skills, and I’ll be looking for any ways to contribute there.

From a code contribution point of view, we have already picked at least two ways we can contribute; things that we think may be useful for the broader community. One of them is a real-time query format translator between MySQL formatted queries – which rely on built-in functions like CONCATENATE and different ways of using LIMIT – and MSSQL server. The other is a more ambitious project, to leverage our exposure to Google Analytics and Google Website Optimizer, and our own internal needs to integrate more closly with the whole Google Apps ecosystem, and contribute the Perl code we write to interface with GData sources back to CPAN. We’ve also now made personal connections with a number of the luminaries in the Australian Open Source community, and with people to talk to about contributing, who can give us access, we’re much more able to play a part.

Stand by for some updates in a month or three as some of these contributions take shape, and thanks again to the organisers who manage to run such a professional, well attended event for what is a seriously (comparatively) low ticket price.

Update: Stephen Edmonds, one of the nicest prolific photographers ever at these sorts of things, took the following photo; some of the Hiive crew with Larry Wall, Father of Perl. Worth the price of admission 😉

Larry Wall and Hiive

Larry Wall and Hiive

… after the jump

I’ve been around blogging for over 3 years, but I’ve always been too busy to contribute actively to the discourse online. I’ve made a few posts here, the most recent getting close to three years ago, and while I’m busier than ever, putting the time aside to collect my thoughts enough to share them – even if it only with the Googlebot – is a discipline I’m hoping to stick to.

So, if you’re reading my blog for the first time, and wonder what the hell happened between early 2006 and now, a couple of weeks from the end of 2008, that’s it: no coma, no abduction by North Korea – just busy and priorities.

Hopefully you enjoy browsing, and feel free to contribute to the discussion: my twitter handle is @geoffmcqueen or you can catch me on Skype (chat preferred to phone calls) at geoffmcqueen.