Cloud Computing Failure

I had a good laugh this week reading about all the consternation about Gmail going down on Monday. If it were only the general public feeling this way I would be more sympathetic, but this is from technologists, geeks even. What's all the fuss about?

A common refrain is that Google is trusted and now that trust is stained. I say baloney. Nothing is certain, and certainly not perfection. One thing I've learned from decades in this business is certainty of availability of communications, computer resources and data is impossible. Sure a system can be made very reliable, which those of us in the business often term high availability, but fail-safe? Nope.

Think of all the things that can go wrong:

• Internet connection goes down or you are out of range of a connection
• ISP or upstream internet congestion or fault, whether caused by act of some god or human
• Server or computer failure, hardware and software
• Fault propagation, where one failure cascades through a network

There's much more. No, the problem isn't that failures occur it is that we are so often unprepared for them. It is far too easy to trust someone else to worry about it for us, even though they can't, and don't, guarantee perfection, and then to whine about it.

Do you keep backups? I know better, yet when I very nearly lost a hard drive recently my backups were not up to snuff. If your internet connection goes down, do you have a ready alternative? Dial-up or even a drive to the nearest Wi-Fi hot spot could serve. If your email is critical do you have a backup? You could auto-forward all email to a separate service, even Gmail, since email data is often difficult to backup and restore. If you use a CRM like that of Salesforce, is your data exportable into a standard format, and do you do so regularly? You should. Backup software inconvenient to use? Find a data buddy, network your computers and then you can conveniently dump your critical files to each other's computers. A USB memory stick also serves nicely.

Murphy's Law says that things will go wrong, especially when you're sure they can't. Services out there in the internet cloud are no exception. Take the trouble to have a disaster plan. Otherwise, the victims ought to do their complaining in front of a mirror.

Major Banks Sitting on 50-day EMA

All the major Canadian banks, which mostly move in sync, with this morning plunge are all sitting at their 50-day exponential moving average. It will be interesting to see if that holds, though it may take the rest of this week to find out.

My bet is that this line holds for at least the short term, or at least is held at this spring's lows. My expectation is that earnings releases in the second half of August will justify higher share prices. This, of course, is no assurance that the market will agree with me!

Wireless Competition Will Be Different This Time

I'm a little late to this story so if you have followed it I'm sure you've read one of more of the many articles about the coming of increased wireless competition in Canada. The reason I'm lagging is I didn't feel I had much to add, though now I see there is some confusion since the various commentators' conclusions span the range from: it'll be great to same old, same old, and everything in between. So just who is likely to win? Will it be consumers, the new entrants, the incumbents, the arms suppliers (equipment vendors), or perhaps everyone?

We know what happened last time. New entrants like Clearnet and Microcell were eventually acquired by the incumbents, which reduced the wireless competition to the former status quo. There were many reasons for this, though in the end it came down to a simple business decision: the companies' stockholders (or bond holders) decided their best financial outcome was to sell rather than remain independent. It was the right decision for them, notwithstanding the dismay of consumers who found their competitive options halved.

Without getting into gory details, I have come up with a list of things that are different this time around. Not necessarily good or bad, just different. I won't pretend the list is complete, and I am not going to even try to quantify the effects of each. Here it is:

• Number Portability: Reduces the friction of switching between service providers. We should expect to see a continuation of obstacles to smoothly-executed number moves, by both new entrants and the incumbents.
• More Than Telephony: Cell phones used to be all about telephony pricing plans and coverage. This time it'll be more than a price war over calling plans. Expect to see data, services and device promotions by the new entrants that will appeal to consumer and business users alike.
• Vendor Financing: This was very prevalent in the 90s by the major vendors, including Lucent, Ericsson and Nortel. That just won't happen this time. The new operators won't be able to share their capital-expenditure risk with this group of companies.
• Investor Financing: Expect to see the bulk of the capital raised coming from equity investors and high-interest bearing bonds. In combination with the lack of vendor financing, the cost of capital will be higher. One consequence is that the new entrants' senior management will be more accountable since those investors will have much greater control. In other words, management will have to perform or they'll be rapidly replaced.
• Crippled Competition: Incumbents like Rogers, Bell and Telus are not as cash rich from telephony (and cable TV) profits as before. Years of direct competition between cable and telcos have whittled margins, forcing them to cut costs. They are less able to sustain price wars than before. While this will fail to drive service prices more than a moderate amount, it will mean the new entrants will slow down the cash burn, and perhaps become cash flow positive in a few years.
• Walled Gardens: The incumbents' stranglehold on what services and devices users can access will be very evident to their users once the new entrants open their data networks and permit new and innovative devices. Expect to see more LiMo, Android and even Symbian devices (maybe even iPhone) available on these new networks, combined with unrestricted internet access and phone applications. Back in the 90s the new entrants used much the same set of phones as the incumbents, but they were higher cost since they could not order the same quantities from the manufacturers. The incumbents will likely be slow to change their service and device strategies, which will give a temporary advantage to the new operators.
• Bundling: The incumbents can bundle like they never have before - wired phone, wireless phone, broadband, TV and long distance. The new entrants have ... mobile wireless. This will increase friction of switching, and may make the new entrants' effective pricing more expensive. They need to find ways to co-market with other companies to at least partially restore the competitive balance. In the present environment good partners will be hard to find.
  

I don't know if an increased number of wireless competitors will be sustainable, which can be difficult in a market the size of Canada. All I know is that the competitive landscape will almost certainly evolve quite differently than it did in the 1990s because the environment is so different.

Gold Miners Feeling Pain

Not long ago I wrote about how gold miners appeared to be dropping more than the underlying commodity. As I saw it then in the charts, the divergence wasn't unusual. Now it is.

Pull up a chart of pretty much any major gold miner, compare it to the commodity, and you'll see what I mean. In the example here I am comparing the gold ETF with Kinross; be sure to use the US stock here to eliminate CAD:USD exchange rate effects.

The chart shows week end (August 8), and today the divergence is growing larger. I am seeing a lot of miners down 25% to 40% over the previous 6 months. Over the same period, gold is trading in a relatively narrow range of -10% to +10%.

Lots of commentators are talking about gold nowadays. One common conclusion they come to is that money is rotating from all commodities into other sectors. I don't know the answer but I am watching closely. I hold one gold miner and I need to decide what to do.

Google Android Delayed and Impaired

When I first wrote about mobile platforms that I was evaluating for business purposes, I concluded that Android might be a suitable first choice even though there are no phones at present and the platform's present incompleteness. I am less certain now.

Delays are to be expected, so even early 2009 for the first real phone is not a deal killer, and remains a target I can live with. Of greater concern is the lack of enthusiasm in the wider developer universe and gaps in the SDK and emulator. Enthusiasm is hard to measure, so let me put that and focus on the platform itself.

The SDK and IDE seem to work fine for me, at least in the very limited use I've made of them. The documentation and teaching material is rudimentary, occasionally incorrect, though generally servicable. There are however two big gaps for my particular purposes:

• Telephony package is incomplete: You can trap key telephony events but you cannot retrieve important data associated with incoming calls, and especially not calling number. The digging I've done indicates that it is coming, sometime, in a future release, though I have no way to know if or when.
• Simulation of calls: Seems to be a challenge to inject calls into the emulator to test telephony applications. I have no idea if this will ever appear in the SDK, or perhaps I must wait for the first phone and (locally) supporting network. This isn't good.

I read elsewhere that a Google marketing manager (Chu) described the Android OS as 80% complete. I hope he didn't mean this in the sense of the software 80/20 rule: that it takes 20% of the effort to complete the first 80%, and 80% of the effort to complete the remaining 20%! Considering the size of the project I am doubting they are serious about launching the first phone by year end.

I get the impression Google is primarily interested in enabling stand-alone apps (those not integrated with telephony) to compete with Apple, RIM and Microsoft, especially to increase distribution of their own apps. Even to accomplish that they have some way to go. I still hope they do succeed since I like the openness of the platform and their comparatively uninhibited approach to outside developers.

Update: From Venture Beat, in all its tedious detail. More waiting for the latest SDK.

49 Year Old Olympian

I don't pay much attention to the Olympics, though some of it does filter through to me regardless of my best intentions. One item that penetrated my wall of disinterest today was this item. It brings a smile to my face to think that a woman about my own age is able to not only compete in the Olympics but to do very well indeed. This was one tough cycling road race and she got close to making the final selection. That's truly impressive, and especially so for someone of a certain age.

This fits well with an earlier post where I suggested there is sense in taking more risks as we get older. Longo-Ciprelli is a good example. Of course it helps to have the genetic make-up to be a world-class athlete, so this particular goal isn't for everyone. Nevertheless I believe the idea is right on the mark, and we all have rewarding goals we can realistically strive for as we grow older. Her example is an encouragement for those of us gradually joining the gray-haired demographic.

Currency Effects Follow-up: Technical Analysis

In my previous post I talked about US currency exchange rate impacts on Canadian investors. I want to add to that by discussing a related issue - technical analysis of stocks that trade on both US and Canadian markets.

Some years back I became concerned when I first became aware that the US and Canadian quoted share prices diverge and converge with the exchange rate (see yesterday's charts). For those that trade in whole or in part on the basis of technical analysis (TA) rather than purely on fundamentals, there is a phenomenon to be aware of. What concerned me at that time is when CAD:USD is volatile the trend lines and support (resistance) levels differ for the stocks on the two markets. The lower the price volatility of the stock, the greater the effect.

Puzzled over how to deal with this I pestered one of the TA commentators on Jim Cramer's Real Money web site (I was a paying subscriber then, but not now). The answer I got did not satisfy me at the time. What it boiled down to was to continue applying the usual TA techniques without regard to exchange rate effects.

As I thought further about what he was telling me I began to better understand his message. TA is not a rigid framework for making money, although it does have value in selecting buy and sell points, whether you're a day trader or a long-term investor. The key to TA is holding to a discipline of buy and sell decision, based on TA indicators, as a superior method for money management. Money management is merely a fancy name for limiting losses from losing trades and protecting gains from winning trades.

On the basis of that understanding I saw that he might very well be right - focus on money management discipline and not on possibly superfluous influences, much as trader emotion is a superfluous, and frequently destructive, influence.

I still have some misgivings about the lesson since it offends something in me to discard information, yet it has not appeared to harm me to do so. You should come to your own conclusion on this lesson since it is your money at stake.

On a related note, it appears that today the CAD:USD rate dropped below support at ~0.94, coincident with commentators in the US noting that the USD has broken out to the upside (versus a basket of currencies, not solely versus CAD). This is good or bad depending on which side your bread is buttered.

Currency Effects on Canadian Investors

The financial world centres on the US dollar. This is of particular importance for those of us who invest in Canadian companies that operate in a global market. Here are some common examples of non-diversified investment vehicles that are impacted:

• Commodities and commodities-based stocks: This includes raw commodities (futures, other derivatives, ETFs) such as silver, zinc and potash, and a whole host of Canadian companies, including Pan American Silver, Lundin Mining, Agrium and Petro Canada.
• Stocks trading on both Canadian and US exchanges: Many companies list on multiple exchanges world-wide to access investors in financial centres where they have a substantial business or investor interest. These range from primarily Canada-focused businesses like Telus and Suncor to multi-nationals like Nortel and Canadian Pacific. There are also companies that started in Canada that list only in the US (Entrust) to those that have found the more onerous listing requirements in the US burdensome (Zarlink).
• Canadian companies reporting in USD: Just as English is increasingly the global language of business, the USD is the currency of choice for companies that operate globally, or aspire to do so. This can be a bit confusing since, while the companies key financial metrics (P/E, revenue, debt, etc.) are priced in USD, the basis of the exchange used for reporting can be obscure without some digging and it typically does not reflect the exchange rate when you buy and sell shares.
• Canadian companies primarily reliant on exports to the US or products priced in USD: This is not only true of all commodities companies but also heavy exporters like Bombardier and Magna. A stronger CAD:USD rate reduces CAD-reported revenue and earnings when the prices of products is maintained in USD.

Small moves in the CAD versus the USD not only matter, they can add up substantially over time. Let’s look at one company, Canadian Natural Resources, which trades on both the TSE and NYSE, reports in USD, and is largely priced on the basis of their primary product, crude oil, which is a commodity priced in USD.

Notice how the price of the two stocks diverge based on the CAD:USD rate. The CAD:USD exchange rate has moved wildly over the previous two years, impacting the prices of the company’s stock on the TSE and NYSE. It is true that the value of a share invested on either exchange remains about the same when measured in one currency (less currency conversion losses in both directions), yet even if you ignore the NYSE entirely the value of your CNQ share varies with CAD:USD.

If you hold an investment for any length of time you need to consider the effect of a trending exchange rate. Even if you're a day trader, the daily volatility in the exchange rate can effect returns; the share price of a dual-listed stock may go in opposite directions on the two exchanges on days when the CAD:USD rate moves significantly.

And, last, if you bought USD$1.00 when the CAD:USD rate was 0.66, that dollar bill has lost 1/3 of its value in CAD. The same goes for a share in a US stock that has the same USD quoted price then and now.

News: Timeliness (2 of 2)

Now that I'm back from the long weekend here is part 2 of my article on news timeliness. Part 1 is here.

For some years now I have used MyYahoo as my main navigation page for markets and stocks. It suits many of my needs despite some shortcomings. I supplement it with a wide variety of other news sources. Right now I'll focus on how MyYahoo meets my criteria for timeliness. If you use it you'll know that they have recently modified their home page design, which both improved and degraded its timeliness performance. That makes it an interesting study for this post. I'll also compare MyYahoo with a few others.

In regard to the market, I listed my priorities in part 1. Let's look at any one particular stock. What I want to know is, the price quote and volume (the basic metrics), what's new, what's recent, and any related news to is relevant to the stock but not about it specifically.

To track stock price and price movement and news I created a variety of portfolios. These are, variously, organized by business sector, how I am monitoring (current holding or with an intent to trade), broad market indices and sector indices (including ETFs and commodity futures).

In the morning, pre-market open in New York and Toronto (9:30 EST/EDT) I want to find any new articles about all the stocks and indices in my portfolios. In the previous version of MyYahoo there was a '*' placed next to each symbol if there was news in the previous 24 hours. This was very helpful. It is no longer there in the new version, much to my dismay. I am now required to touch every symbol or troll through every portfolio and figure it out myself.

When I do touch a symbol or a portfolio within the home page (not navigating to a stock or portfolio page) I am presented with some recent news items, timestamped with minutes, hours or days before present. However if you navigate to a stock page (example) or portfolio page, they continue to give you articles timestamps with the date and time. The latter is what I strongly prefer, and here's why.

Imagine that I touch/open a symbol and it tells me that a company I follow closely or even hold a position issued a press release saying they are pre-announcing upside earnings. This is urgent market-moving news. It then says the news is 15 hours old. This is insufficient. I need to know the exact time of the announcement so that I know: was this after the previous day's market close, during extended-trading hours, and how any price or volume action in response to the news. I now much mentally calculate the time myself or open the article to learn if it contains a better timestamp. For more recent news (where time-before-present is shown in minutes) I must find the base time Yahoo is using, which is elsewhere on the page, and mentally subtract those minutes from that time (not the present time).

At least Yahoo does show detailed timestamps on their stock- and portfolio-specific pages, and I fervently hope they do not change this. Google finance uses these relative timestamps for all their news, which is very inconvenient and one reason I make only limited of use Google finance. One good feature of the new MyYahoo home page is the ability to mouse over a stock to call up the daily price chart; that's nice, even without the volume data.

Stock and index prices are also news. There are 3 types of price: real-time, delayed, and extended hours. Yahoo supports all of them. Unfortunately their real-time and extended hours quotes have degraded to unreliability since they partners with a single quote service (BATS) a few months ago. The reason is that one service sees a fraction of trading activity, resulting in real-time and extended hours quotes that are often missing entirely or indeterminately late for stocks other than those with the highest volume. This is one place where Google surpasses Yahoo nicely, and I use it for that purpose. Unfortunately real-time quotes are only for US stocks; I do not know of any reliable and free real-time quoting services for TSE-listed stocks. If you are a day trader (I'm not) you'll want to subscribe to a good provider of Canadian and US real-time quotes - it is essential news and well worth it.

Let's call up that example Yahoo stock page again. Here we see the news items ordered by timestamp, which is how I like it since I can easily relate the items to price quotes. They also have a rich selection of news sources, including, importantly, common venues for company press releases and financial filings with market regulators. Many of their news feeds are, to my needs, worthless, so I am also happy I can customize them. Another downside of their many feeds is that popular stocks are overflowing with news, making it difficult at times to separate the wheat from the chaff.

Price quotes, which as I mentioned is also news, is continually updated while the page is loaded, both delayed and real-time quotes. Their recently-introduced scrolling stock ticker is useless to me since, like streaming media, I have to sit on my hands and wait for something of interest to appear, which I will not do. In comparison, Google provides a non-scrolling list of recently-quoted stocks on their finance home page, which is updated while the page is loaded.

One curiosity of Yahoo's news feeds is they now have a separate list for blogs. This betrays some confusion on their part as to what is news. It's gets more confusing since many of these blogs items also appear in the news list. Hopefully they'll figure this one out eventually.

Now let's briefly look at Google finance using the same example. As with all Google news properties they use the time before present timestamp method, which I criticized earlier. Their choice of news feeds is small, eschewing many important news items, making it unreliable for stock trading purposes. They do however include some feeds that Yahoo does not, which seems to be particularly better for stocks that trade on overseas markets. Neither Yahoo nor Google uses good Canadian news sources, which makes it dangerous to rely on either for Canadian stocks; we need to supplement both with sources such as Report on Business and the Financial Post, to name two mass media sources.

Google's stock page, like Yahoo's, includes real-time quotes for US markets. These quotes are far better, reflecting the volume and quality of their quotation sources. What they don't do is supply delayed quotes if real-time quotes are available. This is a minor issue since I like to compare real-time and delayed quotes since it gives me a quick indication of how the price is trending.

I have picked on Yahoo and Google since I am most familiar with them and they make suitable examples to illustrate my points about market-related news timeliness. There are many other choices out there, some of which I also use. One gap in my comparison is that I did not include Google's tools for personalizing my finance home page (similar to MyYahoo), though from what I've seen it offers little different from what I discussed above.

Seeking Alpha, which I monitor for stock and market news and analysis, includes some useful news articles and analysis from various blog writers, selected using a modest amount of editorial control. Their list of stocks is, regrettably, not ordered only by time, but also (from what I can glean) by popularity of the articles and of the authors, and the home page shows no timestamp at all. Their sector-specific pages do better in this regard. Market Currents on their home page (and available by RSS feed) and Wall Street Breakfast (also available by email) provide time-sensitive stock and market-moving news during the day and before market open, respectively. Seeking Alpha does include some news on Canadian stocks, from writers who follow Canadian markets, especially (these days) commodities.

Now let's briefly move on to more general news sources.

Even two (vaguely) similar sources of non-urgent, though hopefully important, technology news like Light Reading and Internet News use timestamps differently. On its home page, and on its newsfeeds page, every article has a date. This is not true of Internet News; even the dated box for "Latest News" that show today's date does not match the timestamp for the articles it contains. Before their latest redesign, Internet News did show dates on their home page; now I have no idea of the timeliness of any of their articles, making them a poor source of news.

Mainstream publications, especially the online properties of newspapers, do better. Look at the home pages of The Globe and Mail and the New York Times. At the least the top item in every category has a timestamp, often including time of day, and within categories articles are generally ordered by time. This expands in the same way when you navigate to a category like Report on Business. This is helpful. However, since their redesign (and removal of the annoying 'please pay us' pages) the Economist has eschewed timestamps entirely on their home page. This is disappointing, and certainly irritating. Just how old is their news? I can't tell within a lot of clicking around, which I will not do.

Closer to home there's the Ottawa Citizen. They opt for ordering by time within category (though not always), and their news feed on the right used the elapsed time before present format. Not ideal, but sufficient for a general source of local news. The CBC is an oddity. Especially for local news, apparently news doesn't happen when they're not in the office; you often have to wait until the following business day for news to appear, and even their RSS feed is severely delayed. CBC online is not a good source of timely news.

That's all I want to say about timeliness. When I next talk about news the subject will be relevance.

Sewage Spill Fines - A Better Resolution

Politicians do come up with more and interesting ways to take our money. If you're in Ottawa you know about this sewage spill, and that the city may face substantial fines. Keep in mind that the city's money is our money, raised through property and other taxes.

If the province successfully fines the city we end up paying three times over:

• A fine of several million will have to be recovered in taxes, probably property taxes, since that is a significant hit to a budget where every 0.1% change makes the headlines. Any legal costs incurred in first fighting the fines only add to the bill.
• Don't expect that our tax dollars that are collected by the province by means of fines to the city will be returned to us through income tax reductions or new infrastructure grants to the city. It goes into general revenue to help balance provincial accounts, and the provincial government will crow about being that many million dollars closer to break even.
  
• We need to spend tax dollars on infrastructure to prevent future spill occurrences, and avoid more fines and court cases.
  

None of this should please us; no matter the outcome you and I foot the bill. And remember that it doesn't matter which level of government pays what since there is just the one body of taxpayers.

If all our governments want to demonstrate how effectively they spend our money there are better ways than these tedious and expensive process where we lose and only one of them wins. Settle this quickly and without recourse to the courts. Assess a substantial fine, which the province should then pledge in full as a grant to the city to make the necessary improvements in our sewer system, which the city already indicates will be a priority in the 2009 budget.

This way every level of government can boast about a win, taxpayers get the best value for their money, and we can proceed to protect the environment sooner and most economically.