Monday, December 28, 2020

Some personal anecdotes

People do sometimes ask me about the other part of wine writing, since I usually write about data, rather than about wines themselves. Do I drink the stuff, and what do I think about it? Well, just for a change, I will write one post about it.

My wife and I do consume a bottle or two in most weeks, and we do visit wineries when we are on holiday (see From the sublime to the ridiculous in winery architecture). However, my wife is much better at wine tasting than I am — embarrassingly so. I have the theoretical knowledge, and she has the practical ability. This works well as a team; but I let her write the wine notes, these days.

The current wine shop; Bill sold it in 1986, and opened a much bigger one at Mobbs Hill Cellars

I first learned about wine back in 1982. My girlfriend and I found out that down the road (in the Sydney suburb of Summer Hill) from our rented house was a wine shop, run by a guy named Bill Ireland (a wine retailer in various parts of Sydney from 1979 to 1996). He held tastings of new releases each Thursday evening; and so we became regulars on Thursdays (in those days, Thursday was late-night shopping, when shops stayed open until 9 pm). Bill then started doing special (paid) tastings on Saturdays, which we also attended. I learned as much as I could; and we also started visiting wineries within a day's drive of Sydney (in those days, that meant the Hunter Valley, to the north, and Mudgee, to the west).

Of course, I encountered the usual sorts of problems. For example, I would read a description of a wine and it would bear no connection at all to my own description. Wine writing can be piffle, as I have mentioned before (Are CellarTracker wine scores meaningful?). However, it is all we have as a means of communication, and it is thus not all meaningless. I first learned this at one of Bill's tastings. I had never smelled "chocolate" in a wine, even though the pundits sometimes mentioned it. Then, one day I picked up the bottle from which we were tasting, and sniffed the opening. "Milk chocolate" was the perfect description; and I have never forgotten that experience, in the 40 years since then.

But I still couldn't really get it from the glass. This told me that glasses can also be important, even though some pundits dismiss this idea. I have since found the same thing with some "port" glasses that I have. The contents never seemed to smell of much, and I eventually decided to try some new ones, from a different manufacturer. The difference was impressive, with a whole new world of smells becoming then available in the new glasses.

The Fire Block vineyard, with Bill Ireland on the right

Bill himself was pretty impressive when it came to wine tasting. My favorite anecdote involves a wine from Bowen Estate (Coonawarra, South Australia). Bill used to tell fishing stories about the owner, Doug Bowen, so there was some familiarity. One day, I bought along a special bottle to a Saturday tasting, in lieu of paying. To our great surprise, Bill immediately identified the masked wine — maker, grape, and year (Bowen Estate Claret 1975); and he had never tasted the wine before!

He achieved this apparently remarkable feat by using both experience and deduction. He said that the wine reminded him of the only wine he had tasted of the first vintage from Bowen Estate — there were two reds (Cabernet sauvignon; and Shiraz-Cabernet = Claret), and he had tried the one but not the other. After tasting the masked wine, Bill remembered that he had previously told me this, and he figured that the wine I had brought along would be the missing one. So, it is important to remember your customers, as people!

I have only one anecdote of my own in similar style. I had always read about the smell of "sweaty saddle" in some wines, but had never experienced it. Then, one day, at dinner with a friend (the same one who had told me about Bill's shop in the first place), she asked me whether the wine she was going to serve "smelled okay". Well, one sniff told me that "well-used leather" was the perfect description for this wine's smell. This sweaty-saddle smell had always been described (by Len Evans, the doyen of Australian wine writers) as a characteristic of Shiraz wines from the Hunter Valley. I then remembered my friend telling me that the Hunter was her favorite wine excursion; and, of course, I expected that she would be serving me a special wine. So, I seized my opportunity. I put the glass down without even bothering to taste it, and said: "If it's an old Hunter Valley Shiraz, then it's fine." I then walked away, while she picked her jaw up off the floor. The visitor had got on the wine scoreboard first!

Lewis Perdue (Misinterpretation: Words = Big trouble) notes: "Whether we are aware of them or not, smells do connect to memories that can influence moods". He tells an anecdote about his own unfortunate association with "Almond paste and flying roaches", with which many of us will have empathy.

My own personal version does not involve wine, but involves strawberry flavored milk, instead, which I loved as a child, back in Australia. It was available for purchase in small containers at the school canteen, which I acquired often. Until, that is, my school class went on an overnight train trip from Sydney to Brisbane (1,000 km away), when I was about 11 or 12 years old. This was a slow and bumpy trip, lying in a sleeping berth. I made the mistake of consuming some milk before going to bed. I was queasy during the night; and I have never been able to stomach strawberry flavored milk, ever since. I had to switch to chocolate flavoring, which I still drink, to this day. (Mind you, the chocolate flavoring used here in Sweden is far too sweet to be really nice.)

By the way, Bill Ireland was an aspiring wine-maker of sorts. Even as a retailer, he would buy a few batches of bulk wine, bottle half of it and sell it to us (often under the name Bacchus Selection), and then use the proceeds to put the other half in oak for a while. He would then sell that second half to us at a bigger profit. He and his wife (always known as Noel) eventually purchased a half stake in Finders Bay winery (in Western Australia); and in 1995 they bought what was then Old Station Vineyard (planted in 1926 at Watervale, in the Clare valley, of South Australia), but which they renamed Fire Block in 2003. (See here; or you can use Google Translate to read this Swedish description.)

The wines for the latter are made under contract at the nearby O'Leary Walker winery; and you have never heard of Fire Block, because they are mainly sold locally. Imagine my surprise when, back in late 2007, I came across a 6-pack of Fire Block Old Vine Shiraz in a local liquor store here in Sweden. They were being imported by another expatriate Australian (through Antipodes Premium Wines), who still runs a wine import company (now called Vinoteket). Indeed, to this day Fire Block wines are available only in Australia and Denmark/Sweden (and sometimes Malaysia); and I acquire a suitable selection of each new vintage.

It is not everyone whose connection with their wine mentor still links after 40 years, and literally across two hemispheres! (Update: Bill occasionally contributes items to Australian Rural and Regional News.)

I also had the Bowen Estate Shiraz 2011 with dinner last night. These days, the wines are made by Doug's daughter, Emma (Bowen Estate Shiraz – a Coonawarra benchmark). They are not available in Sweden — I have to order them from the UK.

Monday, December 21, 2020

Ad Tracking by online wine shops is variable

In my previous post, I looked at the extent of Ad Tracking by several dozen websites, covering various parts of the wine industry (Some wine industry websites track their visitors egregiously). I found that only about one-fifth of the sites did no detectable tracking of visitors. More importantly, some sites had outrageous numbers of ad trackers and/or third-party cookies.

I suggested at the time that it might be equally revealing to look at online wine retailers; and that is what I do in this blog post.

As in the previous post, i have used the Blacklight Privacy Inspector to assess each of the nominated websites. This recently released tool is intended to let you find out who is tracking you when you visit their website (see: It’s easier than ever to find out how your favorite websites are tracking you). Blacklight checks for the presence of each of seven possible types of Ad Tracking technology on each site, which are described in detail in an online article (How we built a real-time privacy inspector).

I ran Blacklight across the website of each of the selected online wine sellers last week. These stores are a somewhat arbitrary collection of six dozen possibilities based in the USA (27), the UK (23) , or the European Union (22). Most of the non-US stores are ones I have used myself, at some time over the past decade or two. Sadly, I cannot check any wine retailing that is part of general food stores (such as Amazon, Aldi, Lidl, Costco or Trader Joe’s), because Blacklight checks the main home page of the website, rather than just the subpage specific to wine.

The table at the bottom of this post presents the results of my tests. The columns of the table present each of the seven Blacklight criteria, for each of the named websites (one per row). The first two criteria list the number of trackers encountered, while the other five columns indicate (Y=yes) that the site was detected carrying out the named tracking activity. A summary of Ad Tracking: the 3 main types & how to do it effectively will put you in the picture.

Only 8 / 72 (11%) of the websites were not detected doing any ad tracking at all; and congratulations to those one-ninth of the website owners. Sadly, none of these sites were based in the USA, with five being in the UK and three in the EU.

Indeed, in general, the US sites are doing much more tracking than the other locations. This may have more than a little to do with the much more stringent internet-privacy regulations that have been implemented in the European Union (of which the UK is still a part for the next couple of weeks). The EU collective General Data Protection Regulation differs notably from the state-based approach in the USA (eg. the California Consumer Privacy Act). Actually, the EU is intending to make things even harder in the near future (EU targets big tech in new proposals).

This is not to say, of course, that all US sites are equally bad, because they are obviously not. Indeed, 15 of the 27 sites (55%) were not doing a great deal of tracking; and two sites actually had only one Ad Tracker, while one site had only one Third Party Cookie. However, the other half of the US sites were doing egregious amounts of tracking. The Blacklight check on 80,000 of the most popular websites indicates that the average number of Ad Trackers per site is 7, while the average number of Third Party Cookies is 3. So, those online stores with more than a dozen Trackers (maximum = 27) or Cookies (maximum = 59) are sailing pretty close to the wind.

Most of the UK and EU websites are behaving relatively responsibly, although there are clearly three UK sites that exceed the tracking behavior of the others by a wide margin. Interestingly, one of these is Naked Wines, which has both a US and a UK website — the UK site does notably less tracking that the US site, but still much more than most of the other UK sites.

As far as informing Facebook and Google about your site visits goes, almost all of the US sites did both, while less than a half of the UK and US sites inform Facebook, and less than half of the EU sites inform Google. So, I can expect to receive far fewer personalized ads following me around the web, directly as a result of living in Europe.

In contrast, the EU and UK websites are more likely to be be doing Session Recording than are the US sites; and Keystroke Capturing was detected in the EU but not in the UK. Neither of these practices should be necessary for the purpose of ordering wine online.

To me, the oddest experience occurred at the Drizly store. This was the only site with a CAPTCHA, to assess whether I am a bot or not; but it then proceeded to track me mercilessly, when it realized I am a human (highest number of Cookies, third-highest number of Trackers). I feel doubly insulted!

In my previous post, I specifically noted that many people who have set up their own website may not fully understand what sort of tracking they have allowed on their site (and, indeed, one site author expressed complete surprise, and immediately started doing something about it). However, professionally run wine stores have no such excuse. The offenders listed in the table below need to get their act together.

Note that assessing the US stores is complicated. After all, the infamous Three-tier alcohol-distribution system disrupts the idea of a national online liquor retailer (unlike in the EU, where there is free trade across all member countries). The list of Beer, Wine & Liquor Stores in the US suggests that there were 45,952 liquor businesses back in July, and none of them have a market share of greater than 5%. So, it would be hard to check them all.

As I noted last time, there are some people who apparently don't mind having advertisers looking over their shoulders as they browse the web. However, it goes beyond that. Ad Tracking is like having both mummy and daddy following you everywhere you go, and trying to train you to "do it right". If you do not want this to still be happening to you, then not telling them what you are up to is the obvious first step.

This has been a good year for online alcohol sales, because there is a pandemic, and people are not too keen on going into crowded shops. We are not sure how long this will last (see: Covid online wine boom fizzles out), but I assume that many of you are ordering your Christmas drinks online this year. Now it's time to find out what price you might really be paying — when the product is free, that means you are the product (Don't expose yourself: A guide to online privacy).

Monday, December 14, 2020

Some wine industry websites track their visitors egregiously

Much fuss is made these days about the potential invasion of privacy involved in the act of tracking visitors to websites. In many cases, the visitors are not told about the extent to which they are being tracked by any given website, least of all by the website providers themselves. Indeed, this is so bad that the European Union has legislated against it, which prevents quite a number of US websites from serving EU visitors.

Most of this tracking is allegedly done in the name of "providing a helpful customer experience", but in actual fact this requires very little tracking, if any at all. In practice, most of the tracking is done in the name of advertising — tracking the potential customers' online behavior so that they can be served carefully targeted ads. (Note: this is quite a different thing from other forms of web privacy invasion, such as data security, and leaking of personal information.)

Ad tracking borders on the unethical in many cases, and clearly passes that border in some cases. So, the obvious question is: where does the wine industry stand on this matter, in practice? (What people say, is another matter entirely!) It turns out that we can actually find this out for ourselves.

This is because the topic has been the subject of some practical research, in the sense that there are now tools for you to find out what any given website is doing when you visit it. One of the most ambitious of these is the Blacklight Privacy Inspector. The public release of this tool a few months ago did attract some media attention (eg. 87 percent of websites are tracking you. This new tool will let you run a creepiness check). The Markup company itself ran its own checks on 80,000 of the most popular sites, which provides us with some useful comparison data. One of the most unexpected outcomes was that the site owners themselves often did not know that the homepage set-up tool they had used had inserted trackers into their website (The high privacy cost of a “free” website).

The Blacklight tool checks any web address that you give it, and looks for a number of potential privacy invasions that could be occurring:
  Number of ad trackers
  Number of third-party cookies
  Tracking that evades cookie blockers
  Session recording
  Keystroke capturing
  Facebook pixel
  Google Analytics internet tracking
These are described in detail in an online article (How we built a real-time privacy inspector), and I briefly discuss them below.

So, let's name names. The table at the bottom of this post presents a somewhat arbitrary collection of five dozen wine-related websites. I have tried to include all of the usual suspects, supplemented by a few sites that I sometimes read myself. I apologize in advance to those of you who have been left out, or if anyone finds releasing this information embarrassing. My ultimate aim is to make wine-industry websites as clean as possible, by pointing out where the dirt has accumulated.

Note, also, that I have excluded online stores selling wine and accessories, which are far and away the most likely sites to be tracking you (for their own purposes). They may appear in a future post (now online as: Ad Tracking by online wine shops is variable).

The columns of the table present each of the seven Blacklight criteria, evaluated by me during this past week, for each of the named sites (one per row). The first two criteria list the number of trackers encountered, while the other five columns indicate (Y=yes) that the site was detected carrying out the named tracking activity.

I am pleased to report that 11 / 60 (18%) of the sites were not detected doing any ad tracking at all. Kudos to those one-fifth of the website owners. Note that the Wine Gourd does have one cookie, as do all Blogger sites (owned by Google), and also Facebook sites.

Potentially the most invasive activity is Keystroke Capturing, which records every single time you press a keyboard key (such as when entering your personal details in a form). There were 2 sites apparently doing this, which Blacklight reports as an average percentage across the web. It is possible that this use is (or once was) actually legitimate, as it is sometimes used for auto-completion of forms. This would need to be checked.

Session Recording keeps a record of your on-site activity, including mouse movements, clicks and page scrolling. This is sometimes ostensibly done to optimize the layout of web pages, but it can also be used for ad tracking (see Advertisers can now tell if we're paying attention). There were 4 sites doing this, which is half the Blacklight average; but why do it at all? These sites definitely do not need to be doing this.

Evading Cookie Blockers (aka Canvas Fingerprinting) is the next most egregious action. There are browser add-ons or extensions that we can use to stop cookie tracking by any website that we visit (the one I use is Privacy Badger); and when we use them we mean: Don't Track Me! There were 5 websites found to be using fingerprinting, which is a bit above average, according to Blacklight. It is possible, but unlikely, that this activity was actually an attempt to protect those sites from botnet attacks.

Next, Ad Trackers try to identify and collate information about users, and send that information back to the tracker's website, to be used for ad targeting. The average number per website, as reported by Blacklight, is 7, and the wine sites had an average of 6. So, the wine industry is no worse than elsewhere, in this regard. However, you can see in the table that several wine websites currently exceed this number massively. This occurred in all four categories of website, although the Wine Professionals were the least egregious. But why on earth does anyone need to put 20 or 30 ad trackers on their website? Sadly, we all know the answer.

The same thing applies to Third Party Cookies. These have nothing to do with the running of the website itself (which legitimately uses cookies to keep a track of things like visited pages and log-in details). Instead, these other cookies are installed on behalf of external sites, and store information on your own device, to identify you when you visit other sites with the same cookie. Blacklight says that the web average is 3 per site (often associated with the ads displayed on the pages), but the wine industry apparently thinks that a better number is 11 per site. More to the point, you can see for yourselves that quite a number of sites massively exceed even this large average — and there are not that many ads displayed on their pages, so the cookies are covert. These sites are usually the same ones that have a large number of ad trackers, as well; so, we are left in no doubt about the attitude towards site visitors by these website owners.

Next is Facebook, which is monetized by serving personalized ads (and is currently embroiled in anti-trust lawsuits — Federal government and 46 states file antitrust suit seeking to split up Facebook). To serve these ads, Facebook wants to know which websites you visit, and it finds this out by getting the website owners to tell it. Blacklight says that about 30% of sites agree to do this, and 23% of the wine industry sites also currently do so. Did you know that a quarter of the websites you visit are telling Facebook that you are there? This explains why Facebook ads seem so directly relevant to your past web activity.

Finally, we come to Google, which is also monetized by ads, and whose web behavior has resulted in fines of tens of millions of dollars, especially in the European Union (eg. Google, Amazon fined $163 million as France takes hard line on privacy). Google wants to track you across all of the web, which it does by cross-collating every bit of information it can find about you, and especially your activity. In order to succeed, it helps if the website owners tell them about your activity, which Blacklight says about 50% of sites agree to do. In the wine industry, it is about 45% of the websites. So, Google must know a lot about you all, by now.

As I noted earlier, it is possible that some the website owners fingered here do not know what their own website is doing. If they used a "free" package to set up their sites, then the tracking activity could have been installed without their knowledge. Blacklight reports that a number of site owners immediately removed the trackers when they were informed about them. It would be nice if that happened here, as well.

For the rest of you site owners, please clean up your act. A cookie or two is one thing, but the table below indicates some seriously invasive activity. If a site really does need to do this sort of thing, then it should tell us about it when we first visit the site, rather than making us find out for ourselves. Indeed, this is actually mandatory in the EU — although I doubt that too many people read all of those Cookie Consent pop-ups (they are usually quite uninformative).

In the meantime, the rest of you can all try out Blacklight, and test your favorite websites for yourselves. Some people apparently don't mind having advertisers looking over their shoulders as they browse the web. Indeed, anyone who logs into a website using their Facebook or Google credentials is actually demanding that they be tracked (in exchange for saving a few seconds, or having to remember only one set of login details). However, we should all be given the option to opt into tracking if we want to, rather than being forced to opt out.

Wine industry website ad tracking.

Monday, December 7, 2020

Tasting great wines?

A few weeks ago, Eric Asimov noted that: "Benchmark bottles were always a splurge. But an increasing concentration of wealth has put them out of reach for all but the richest connoisseurs" (How income inequality has erased your chance to drink the great wines).

The basic argument here is that The Great Wines of the World used to be at least within the reach of the likes of you and I, even if we chose not to reach out, and actually pay the money. That is, we had a choice: "middle-class wine lovers could still afford to experience ... drinking a truly great wine, not simply to enjoy it, but to understand what qualities made it exceptional in the eyes of history." These days, this choice is no longer the case, so that Millennials will not be able to experience fine wines the way their Baby Boomer parents are alleged to have done. (See my post on: The outrageous prices of modern high-quality wines)

Well, I grew up in Australia. In Australia, even the Baby Boomers found it difficult — we could get good Australian wines, but the stuff from elsewhere was financially out of bounds, even back in the early 1980s. I got thoroughly sick of hearing about them, especially from a bunch of wine writers who usually got to taste them for free. Since I never got to experience these things, my daughter (a Millennial) and I (a Boomer) have more in common than you thought.

Far more importantly, though, I feel that this is all beside the point. Indeed, it smacks of wine snobbery. Wealth has always been the arbiter as to what are The Great Wines of the World. Are these wines expensive because they are great, or great because they are expensive? Neither I nor the Millennials will ever find out.

Obviously, I have had a long time to think about this, and work out what to do about my wine education. I could move down-market (Can’t afford the finest wines from Burgundy and Bordeaux? Try these instead), but that somewhat defeats the purpose. So, I decided that a far more useful yardstick of wine quality is likely to be modern comparative tastings. There are plenty of them; and they often seem to involve wines that we could all afford. Why not get the experience of drinking some of them, and find out why they did so well in head-to-head competition?

The answer to that question is, of course, that many of these tastings are not as objective as we might like. Indeed, some of them are downright suspicious — you pay your entry fee and you get your award. So, we need to tread carefully. In this regard, it has been said that: "The International Wine Challenge [is] the world’s most influential and rigorously judged wine competition". So, we could do worse than start by having a look at the IWC winners.

Conveniently, the IWC has recently announced its 2020 best wines from around the world. The full list of results is on their website, but the top 30 wines have been reported around the media (eg. here and here).

What I have done for these wines is use Wine-Searcher (and some other sources, when that failed) to collate the average $US price per standard bottle, excluding sales tax. The results for 29 out of the 30 wines are shown in this graph (the Chinese wine proved to be elusive). Each dot represents the price of one wine.

Prices of IWC wine winners in USD ex. tax

These prices are not as bad as they might be, although the increase in price across the wines is exponential (as expected). Clearly, some of them are not much better priced than The Great Wines, but others are very affordable. Indeed, a bit of a splurge (up to $100) would get you almost all of them.

So, my suggestion is that the Millennials can approach their wine education this way, rather than by adopting the ways of their forefathers. There is nothing either remarkable or new about this suggestion, but it seems to need to be pointed out again. The Old Way is not always the best way.

The ultimate objective of wine education, as I see it, is well expressed by Tom Maresca (Some good everyday wines): "I long ago decided that life is too short to ever drink mediocre wine, so even though I could never afford those legendary, crème de la crème bottles that headline so many ads, I’ve worked hard to ensure that the wines that accompany my daily bread are pleasurable".