Wednesday, October 24, 2007
PLCB versus the LCBO
Recently, when the US and Canadian dollars reached parity, and 1 USD = 1 CAD, I took my first trip to Toronto.
While I was there, I decided to visit a wine store, to search for a couple of Canadian wines, which I rarely see in the US (at least in PA). Like Pennsylvania, Ontario has a liquor control board, known as the LCBO, through which all wine has to be purchased. According to Wikipedia, the LCBO is the largest and the PLCB is the second largest wine and liquor purchasers in the world. It seemed essential that I compare the two.
The first thing I noticed entering a LCBO store is that they're very pretty. They
make you comfortable with a decor of soothing yellows and greens. LCBO apparently puts a little more effort into decorating their stores than the PLCB. Like the PLCB, the LCBO features several different kinds of stores, ranging from cramped stores with small selection downtown, to larger stores with more to offer outside of the city center (where retail space appears cheaper).
I have to mention it: the employees I encountered a the two Toronto LCBO stores I visited were amazing. First and foremost, they were almost all friendly and outgoing, even though both stores were SWAMPED with customers. Second, they were unbelievably fast at their jobs --- especially the cashiers. Once we got to the cash register, it took about 30 seconds to ring up, pay for, and bag our 3 bottles of wine. I've never seen a cashier move as fast as I did that day. I have never been to a PLCB store and had the same experience.
The LCBO selection is clearly biased towards favoring Canadian wines wherever possible (although this could have been due to close proximity to Canadian
wine producing regions). At least 15 percent of their retail space (and perhaps
more) was devoted to VQA Canadian wines. I have never looked very hard for Canadian wines at the PLCB, but I have only seen a single lonely offering mixed into the "Other Red Wines" section.
According to Wikipedia, the LCBO has a reputation for high prices. I used the online web sites for both the PLCB and the LCBO to generate this small sampling of comparison prices. It was particularly difficult finding the exact same version of popular wines in both stores. I suspect this is due to bulk purchasing done by both organizations --- in 2003 PLCB negotiates the best deal and gets most of Kim Crawfords Selection X, and in 2004 the LCBO gets it.
Even when you factor in sales taxes --- 18 percent for the PLCB and 12 percent for the LCBO, almost all of these wines are generally more expensive in Toronto (and sometimes a lot more expensive) than they are in Pittsburgh. Only four cases in the above selections do the LCBO prices beat PLCB prices. Interestingly, it seems that the LCBO charges more than the PLCB even for many super expensive wines for which I doubt either purchaser is getting much of a volume discount. It would be interesting to see how much profit each organization makes off of each wine.
I have to observe that it's possible that the LCBO prices above may seem high because of the recent shifts in the value of the canadian dollar relative to the US dollar. If we were comparing these prices a year ago, we would have to knock at least 15 percent off the values of the LCBO wines. That makes many of the LCBO offerings look as good as or better than the PLCB offerings. It will be interesting to observe how the LCBO wine prices and PLCB wine prices adjust over the next year or so. My suspicion is that PLCB will have a much more difficult time of procuring foreign wines with such a weak US dollar, but it is hard to say.
While searching online for PLCB and LCBO offerings, I found that the PLCB almost always offers a greater selection than the LCBO. For example, the PLCB offers
59 St Emilions to LCBO's 10, 126 Amarones to LCBO's 66, etc. This, of course, includes wines that have to be special-ordered from the PLCB. If you remove the special-orders from the PLCB listings, you approach numbers much closer to LCBO's. The LCBO, also seems to be a little more liberal about letting you ship wines in yourself, so this may or may not be a significant issue.
While searching the PLCB and LCBO pages for wines to price compare, I found that the PLCB seems to almost always offer a wider selection than the PLCB. This, of course, includes wines that are only offered by the PLCB under special orders
As I mentioned above, I used both the PLCB and the LCBO web pages for a wee bit to research this article. Like their stores, the LCBO web page is about 100 times prettier than the PLCB offering. The functionality, however, is more or less similar. It seems as if the LCBO suffers from the same braindead "show me the wines in my local store" functionality as the PLCB --- You can do it, but it is a pain, taking much more effort than needed. Finally, the LCBO has found a unique way of sucking my soul through my ears, by breaking my search every time that I accidentally hit my "Return" key to submit a form instead of clicking on the "Search" button (It manages to return 0 matches for whatever I search for in that case).
In general, I've found the LCBO experience to be generally similar to the PLCB experience. The prices are possibly a bit higher than the PLCB's, but in exchange you get a cheerier atmosphere, and what appears to be faster service. Still, neither one can really compare to the customer service experiences I've gotten at independently-run wine stores in Ohio and New Jersey. I still don't understand fully why those success stories cannot be recreated inside Pittsburgh and Toronto.
While I was there, I decided to visit a wine store, to search for a couple of Canadian wines, which I rarely see in the US (at least in PA). Like Pennsylvania, Ontario has a liquor control board, known as the LCBO, through which all wine has to be purchased. According to Wikipedia, the LCBO is the largest and the PLCB is the second largest wine and liquor purchasers in the world. It seemed essential that I compare the two.
The first thing I noticed entering a LCBO store is that they're very pretty. They
make you comfortable with a decor of soothing yellows and greens. LCBO apparently puts a little more effort into decorating their stores than the PLCB. Like the PLCB, the LCBO features several different kinds of stores, ranging from cramped stores with small selection downtown, to larger stores with more to offer outside of the city center (where retail space appears cheaper).
I have to mention it: the employees I encountered a the two Toronto LCBO stores I visited were amazing. First and foremost, they were almost all friendly and outgoing, even though both stores were SWAMPED with customers. Second, they were unbelievably fast at their jobs --- especially the cashiers. Once we got to the cash register, it took about 30 seconds to ring up, pay for, and bag our 3 bottles of wine. I've never seen a cashier move as fast as I did that day. I have never been to a PLCB store and had the same experience.
The LCBO selection is clearly biased towards favoring Canadian wines wherever possible (although this could have been due to close proximity to Canadian
wine producing regions). At least 15 percent of their retail space (and perhaps
more) was devoted to VQA Canadian wines. I have never looked very hard for Canadian wines at the PLCB, but I have only seen a single lonely offering mixed into the "Other Red Wines" section.
According to Wikipedia, the LCBO has a reputation for high prices. I used the online web sites for both the PLCB and the LCBO to generate this small sampling of comparison prices. It was particularly difficult finding the exact same version of popular wines in both stores. I suspect this is due to bulk purchasing done by both organizations --- in 2003 PLCB negotiates the best deal and gets most of Kim Crawfords Selection X, and in 2004 the LCBO gets it.
| Wine | PLCB | LCBO |
|---|---|---|
| Twin Fin Cab Sauvignon | 6.99 | 14.30 |
| Fish Eye Pinot Grigio 2006 | 7.99 | 10.15 |
| Louis Jadot Beaujolais Villages 2006 | 10.99 | 16.95 |
| Chateau Cardonne Medoc | 23.09 | 26.95 |
| Two Hands Gnarly Dudes Shiraz 2002 | 29.99 | 24.95 |
| Alkoomi Blackbutt 2002 | 35.99 | 27.00 |
| Leeuwin Shiraz Art Series 2003 | 32.99 | 35.15 |
| Grant Burge Holy Trinity GSM 2002 | 37.95 | 35.39 |
| Two Hands Lily' Garden Shiraz 2004 | 43.99 | 65.00 |
| Corino Barolo Arborina | 52.99 | 66.00 |
| Clarendon Hills Sandown Cab Sauv 2004 | 64.00 | 68.00 |
| Silver Oak Cab Sauv 2002 | 99.99 | 82.95 |
| Opus One 2003 | 164.00 | 143.95 |
| Opus One 2004 | 175.89 | 269.95 |
| Chateau Margaux 1998 | 347.00 | 395.00 |
Even when you factor in sales taxes --- 18 percent for the PLCB and 12 percent for the LCBO, almost all of these wines are generally more expensive in Toronto (and sometimes a lot more expensive) than they are in Pittsburgh. Only four cases in the above selections do the LCBO prices beat PLCB prices. Interestingly, it seems that the LCBO charges more than the PLCB even for many super expensive wines for which I doubt either purchaser is getting much of a volume discount. It would be interesting to see how much profit each organization makes off of each wine.
I have to observe that it's possible that the LCBO prices above may seem high because of the recent shifts in the value of the canadian dollar relative to the US dollar. If we were comparing these prices a year ago, we would have to knock at least 15 percent off the values of the LCBO wines. That makes many of the LCBO offerings look as good as or better than the PLCB offerings. It will be interesting to observe how the LCBO wine prices and PLCB wine prices adjust over the next year or so. My suspicion is that PLCB will have a much more difficult time of procuring foreign wines with such a weak US dollar, but it is hard to say.
While searching online for PLCB and LCBO offerings, I found that the PLCB almost always offers a greater selection than the LCBO. For example, the PLCB offers
59 St Emilions to LCBO's 10, 126 Amarones to LCBO's 66, etc. This, of course, includes wines that have to be special-ordered from the PLCB. If you remove the special-orders from the PLCB listings, you approach numbers much closer to LCBO's. The LCBO, also seems to be a little more liberal about letting you ship wines in yourself, so this may or may not be a significant issue.
While searching the PLCB and LCBO pages for wines to price compare, I found that the PLCB seems to almost always offer a wider selection than the PLCB. This, of course, includes wines that are only offered by the PLCB under special orders
As I mentioned above, I used both the PLCB and the LCBO web pages for a wee bit to research this article. Like their stores, the LCBO web page is about 100 times prettier than the PLCB offering. The functionality, however, is more or less similar. It seems as if the LCBO suffers from the same braindead "show me the wines in my local store" functionality as the PLCB --- You can do it, but it is a pain, taking much more effort than needed. Finally, the LCBO has found a unique way of sucking my soul through my ears, by breaking my search every time that I accidentally hit my "Return" key to submit a form instead of clicking on the "Search" button (It manages to return 0 matches for whatever I search for in that case).
In general, I've found the LCBO experience to be generally similar to the PLCB experience. The prices are possibly a bit higher than the PLCB's, but in exchange you get a cheerier atmosphere, and what appears to be faster service. Still, neither one can really compare to the customer service experiences I've gotten at independently-run wine stores in Ohio and New Jersey. I still don't understand fully why those success stories cannot be recreated inside Pittsburgh and Toronto.
Friday, August 17, 2007
Searching the PLCB
Most people I know tend to think that the Pennsylvania Liquor Control Board (PLCB) provides nothing but disservice to Pennsylvania wine drinkers, and that neighboring states like Ohio and New Jersey have it better with privately-run wine stores.
While I think there's a great deal of truth to that claim, I there is one way the PLCB gives us a clear benefit: The Product Search Page. Well, at least, it helps us in principle.
The idea is simple. Since all of your wine and spirits purchases have to be made through one company, it is feasible to search all the wine stores in the state for a product you want. If it's there, you can run out to the store and grab what you want.
Sounds like a great idea. Those states with hundreds of independent private retailers have no way to even begin address the technical and political complexities in hooking up all their inventories at a one stop shop. You'd need to search dozens of websites, or worse, visit actual stores to find a product you want.
Sadly, The Product Search Page has been clinically shown to cause users to purchase firearms, and shoot themselves in the head, repeatedly, in order to relieve the pain it causes.
I believe the central problem is that the developers had no idea, or didn't care what people want to do with that web page.
Historically, this used to be a huge problem in computer science. Early on, computer programmers wrote programs for themselves to use. They solved the programmers' problems, but were arcane and impossible for anybody else to use. Eventually, computer programmers were hired to write programs for other people to use. Since computer programmers (like me) are geeks who cannot interact with anybody else, and nobody knew better, they wrote programs that only computer programmers would understand, and then taught the users how to use them. "Want to copy some text? Sure. Just hit Control-F1-c! See how intuitive it is?!" Even worse, sometimes those programs didn't even solve the actual problems that users had. Eventually Apple came around and started designing their programs so they're actually intuitive and easier to use for normal people. It was the dawn of a new era.
To be clear, The Product Search Page is from before that era.
What went wrong with The Product Search Page is that the PLCB developed a program without understanding at all what people would use it for. It appears as if they knew they had a database of inventory, and decided "We should put this on the web! We'll be so kewl!" So they created a couple of database queries, put it on the web, and they were done. Instant profit!
In comparison, what should happen in the new era is this: Study (potential) users to understand what they need to do. Develop a user interface to meet users' needs. Go back and study how users use your system and their needs change. Repeat. All of these steps are clearly missing from or deficient in The Product Search Page development process.
The Product Search Page gives you two search queries. First, "Is a wine in any PLCB store or the PLCB special-order system?" Only after I have submitted this query, I can click on a wine and submit a second query, "Is this wine in a given store, or any store in my city?"
I believe that all users of The Product Search Page want to know exactly one thing: If a particular product is available in a local wine store, or a particular wine store. Fortunately, I believe out of sheer luck, I can answer my query using the arcane queries I am given:
Things get even worse if I am just interested in "What Merlot are located at my store?" Then I have this contorted sequence of steps:
Sadly, I cannot even begin to imagine what questions the designers of the system thought users wanted to answer. Once I've done a search, and see that a wine is available somewhere in Pennsylvania, I cannot figure out where --- I can only search for my city and pray that I've guessed correctly. Furthermore, if I see that a wine can be special-ordered, there is absolutely no way to do so via the web.
For those of you who have been using computers anytime after 1984 knows that The Right Solution looks more like:
The most frustrating thing is that it would take 5 minutes of programming to make The Product Search Page do what I (the users) want. The second most frustrating thing is that although it takes a mind-numbing amount of effort for the users, the website can do what I want; Thanks to the PLCB monopoly there is no reason or incentive for the PLCB to even think about fixing this.
While I think there's a great deal of truth to that claim, I there is one way the PLCB gives us a clear benefit: The Product Search Page. Well, at least, it helps us in principle.
The idea is simple. Since all of your wine and spirits purchases have to be made through one company, it is feasible to search all the wine stores in the state for a product you want. If it's there, you can run out to the store and grab what you want.
Sounds like a great idea. Those states with hundreds of independent private retailers have no way to even begin address the technical and political complexities in hooking up all their inventories at a one stop shop. You'd need to search dozens of websites, or worse, visit actual stores to find a product you want.
Sadly, The Product Search Page has been clinically shown to cause users to purchase firearms, and shoot themselves in the head, repeatedly, in order to relieve the pain it causes.
I believe the central problem is that the developers had no idea, or didn't care what people want to do with that web page.
Historically, this used to be a huge problem in computer science. Early on, computer programmers wrote programs for themselves to use. They solved the programmers' problems, but were arcane and impossible for anybody else to use. Eventually, computer programmers were hired to write programs for other people to use. Since computer programmers (like me) are geeks who cannot interact with anybody else, and nobody knew better, they wrote programs that only computer programmers would understand, and then taught the users how to use them. "Want to copy some text? Sure. Just hit Control-F1-c! See how intuitive it is?!" Even worse, sometimes those programs didn't even solve the actual problems that users had. Eventually Apple came around and started designing their programs so they're actually intuitive and easier to use for normal people. It was the dawn of a new era.
To be clear, The Product Search Page is from before that era.
What went wrong with The Product Search Page is that the PLCB developed a program without understanding at all what people would use it for. It appears as if they knew they had a database of inventory, and decided "We should put this on the web! We'll be so kewl!" So they created a couple of database queries, put it on the web, and they were done. Instant profit!
In comparison, what should happen in the new era is this: Study (potential) users to understand what they need to do. Develop a user interface to meet users' needs. Go back and study how users use your system and their needs change. Repeat. All of these steps are clearly missing from or deficient in The Product Search Page development process.
The Product Search Page gives you two search queries. First, "Is a wine in any PLCB store or the PLCB special-order system?" Only after I have submitted this query, I can click on a wine and submit a second query, "Is this wine in a given store, or any store in my city?"
I believe that all users of The Product Search Page want to know exactly one thing: If a particular product is available in a local wine store, or a particular wine store. Fortunately, I believe out of sheer luck, I can answer my query using the arcane queries I am given:
- Type the product name
- Click Search
- Find the product in a big list
- Click on the product
- Type in "Pittsburgh"
- Click Submit
- Find that there are no bottles anywhere near me.
Things get even worse if I am just interested in "What Merlot are located at my store?" Then I have this contorted sequence of steps:
- Type in Merlot
- Click Search
- Find the product in a big list
- Click on the product
- Type in "Pittsburgh"
- Click Submit
- Find that there are no bottles anywhere near me.
- Click Back TWICE
- Repeat steps 4-8 ad infinitum
Sadly, I cannot even begin to imagine what questions the designers of the system thought users wanted to answer. Once I've done a search, and see that a wine is available somewhere in Pennsylvania, I cannot figure out where --- I can only search for my city and pray that I've guessed correctly. Furthermore, if I see that a wine can be special-ordered, there is absolutely no way to do so via the web.
For those of you who have been using computers anytime after 1984 knows that The Right Solution looks more like:
- Type in the product name/keyword
- Type in my store city or number
- Click Search
- See the wine available in my local store!
The most frustrating thing is that it would take 5 minutes of programming to make The Product Search Page do what I (the users) want. The second most frustrating thing is that although it takes a mind-numbing amount of effort for the users, the website can do what I want; Thanks to the PLCB monopoly there is no reason or incentive for the PLCB to even think about fixing this.
Saturday, July 28, 2007
Bona Terra
While searching for new restaurants in Pittsburgh, I happened across Bona Terra, in Sharpsburg. Being vegetarian, I am always leery of new places, so I called to see if they had anything for me to eat. They said that as long as we called a day ahead, they could make sure to have something delicious. I was reassured, and vowed to get there soon. I recently found the opportunity to go while I was moving apartments. I made a reservation (which is, apparently, necessary, if you expect to eat there) for 8pm on a Saturday night.
The restaurant itself is impressively difficult to find, but we called the restaurant from the road, and they helped us find it. From the outside, it looks unassuming but charming.
From a vegetarian's perspective, my first time here was unsettling. Nothing on the menu was vegetarian-friendly (for a non-fish vegetarian), but the wait staff kept reassuring me that anything on the menu could be made vegetarian. I didn't, however, understand how a "Sirloin" or "Filet of Sole" could be made vegetarian. When it came time to order, the waiter recommended that I just ask for the chef to "do his vegetarian thing," and I went with it. I was asked about my preferences for sauces and items on the menu, and was left without a clue as to what was coming.
My girlfriend and her mom had a much easier time, ordering a filet of sole and a sirloin steak, respectively.
For appetizers, we ordered a communal gnocchi (without crab to please me), and a soup and salad. When they came, I got to the gnocchi first. I cannot say this any more simply, but this dish was sublime. One of the most amazing things to pass my lips in a long time. The sauce was fresh and sweet and piquant, and the gnocchi was soft, velvety and luscious. I don't know what Faustian bargain the chef has made to cook this stuff, but I am glad he did.
Subsequently, we were served an amuse and some bread. The amuse was a little peach slice with raspberry coulis, mint, and some blue cheese. It was quite refreshing, but nothing spectacular. The bread was a rosemary-infused challah, I believe, which was tasty, but a bit weak on the rosemary. During this period, we suffered a pregnant pause in the arrival of food, which may have been unintended, but we did not mind, as we were tired from moving, and enjoying the discussion.
Eventually, the entrees arrived. The chef decided to make me some polenta triangles, covered in an olive-relish, and drizzled with a basil oil. Alongside, it came with an assortment of sauteed veggies, including some beans and baby bok choy. The baby bok choy was a bit oily for my taste, but the rest of the dish was, once again, sublime. The polenta was light and creamy and unlike any polenta I've made before (perhaps a high cream content is the secret?). The beans and the olive-relish were beyond-heavenly, and again, unlike anything I've ever cooked. I have something new to strive for. Although I did not partake, my girlfriend and her mom thoroughly enjoyed their meals as well.
In gustatory Elysia, we decided to continue with dessert. We ordered two desserts --- I believe they were a chocolate torte with hazelnut ice cream, and a strawberry parfait with peach ice cream. Unlike my companions, I had prepared myself for this moment throughout the evening, by pacing myself and not stuffing myself on bread. Thus, I was prepared to eat quite a bit of both desserts. The parfait was quite refreshing, although when I go for dessert, I want it rich and decadent. The torte, rich and decadent, was more my style. Sadly, I thought that the torte and the ice cream did not complement each other well --- both were too dark and chocolatey to accentuate each other. I thought that the peach ice cream from the parfait would have been a much better foil for the torte, and ended up spooning from both dishes.
The waitstaff at Bona Terra are pretty much exceptional. They were warm, friendly, and welcoming throughout the entire meal. The service is a little formal --- our waiter continually refilled our wine and water glasses throughout the meal, and presented us with tiny silverware just before they were necessary. For a special occasion, however, it makes one feel quite rich.
It is fortunate that the service makes one feel rich, since the prices at Bona Terra are not cheap. Our dinner for 3 ended up costing somewhere around 150$, which is definitely no bargain. I daresay, however, that I do not regret spending a dollar of that tab: the service, and more importantly, the food was well-worth it. I can't wait to go back for my next special occasion, if only for the gnocchi.
The restaurant itself is impressively difficult to find, but we called the restaurant from the road, and they helped us find it. From the outside, it looks unassuming but charming.
From a vegetarian's perspective, my first time here was unsettling. Nothing on the menu was vegetarian-friendly (for a non-fish vegetarian), but the wait staff kept reassuring me that anything on the menu could be made vegetarian. I didn't, however, understand how a "Sirloin" or "Filet of Sole" could be made vegetarian. When it came time to order, the waiter recommended that I just ask for the chef to "do his vegetarian thing," and I went with it. I was asked about my preferences for sauces and items on the menu, and was left without a clue as to what was coming.
My girlfriend and her mom had a much easier time, ordering a filet of sole and a sirloin steak, respectively.
For appetizers, we ordered a communal gnocchi (without crab to please me), and a soup and salad. When they came, I got to the gnocchi first. I cannot say this any more simply, but this dish was sublime. One of the most amazing things to pass my lips in a long time. The sauce was fresh and sweet and piquant, and the gnocchi was soft, velvety and luscious. I don't know what Faustian bargain the chef has made to cook this stuff, but I am glad he did.
Subsequently, we were served an amuse and some bread. The amuse was a little peach slice with raspberry coulis, mint, and some blue cheese. It was quite refreshing, but nothing spectacular. The bread was a rosemary-infused challah, I believe, which was tasty, but a bit weak on the rosemary. During this period, we suffered a pregnant pause in the arrival of food, which may have been unintended, but we did not mind, as we were tired from moving, and enjoying the discussion.
Eventually, the entrees arrived. The chef decided to make me some polenta triangles, covered in an olive-relish, and drizzled with a basil oil. Alongside, it came with an assortment of sauteed veggies, including some beans and baby bok choy. The baby bok choy was a bit oily for my taste, but the rest of the dish was, once again, sublime. The polenta was light and creamy and unlike any polenta I've made before (perhaps a high cream content is the secret?). The beans and the olive-relish were beyond-heavenly, and again, unlike anything I've ever cooked. I have something new to strive for. Although I did not partake, my girlfriend and her mom thoroughly enjoyed their meals as well.
In gustatory Elysia, we decided to continue with dessert. We ordered two desserts --- I believe they were a chocolate torte with hazelnut ice cream, and a strawberry parfait with peach ice cream. Unlike my companions, I had prepared myself for this moment throughout the evening, by pacing myself and not stuffing myself on bread. Thus, I was prepared to eat quite a bit of both desserts. The parfait was quite refreshing, although when I go for dessert, I want it rich and decadent. The torte, rich and decadent, was more my style. Sadly, I thought that the torte and the ice cream did not complement each other well --- both were too dark and chocolatey to accentuate each other. I thought that the peach ice cream from the parfait would have been a much better foil for the torte, and ended up spooning from both dishes.
The waitstaff at Bona Terra are pretty much exceptional. They were warm, friendly, and welcoming throughout the entire meal. The service is a little formal --- our waiter continually refilled our wine and water glasses throughout the meal, and presented us with tiny silverware just before they were necessary. For a special occasion, however, it makes one feel quite rich.
It is fortunate that the service makes one feel rich, since the prices at Bona Terra are not cheap. Our dinner for 3 ended up costing somewhere around 150$, which is definitely no bargain. I daresay, however, that I do not regret spending a dollar of that tab: the service, and more importantly, the food was well-worth it. I can't wait to go back for my next special occasion, if only for the gnocchi.
Sunday, July 08, 2007
The Light in the PLCB
Yesterday, I found myself in the PLCB store in Robinson Towne Center. It takes me about 30 minutes to drive there, and I pass by no fewer than 10 PLCB stores to get there.
I stopped by since I was in the area anyway, and I needed some Wine Preserver. Wine Preserver is an ingenious little bottle of nitrogen and noble gases that one sprays into opened wine bottles to keep them from oxidizing and going bad. It works a lot better than those vacuum stoppers.
This trip was particularly notable since it gave me my very first pleasant experience at a PLCB store. The staff were extraordinarily helpful, friendly, and knowledgeable. Even though the store didn't have any Wine Preserver in stock, the store manager (I believe) chatted with me for about 10 minutes as he helped me figure out where to get some. He answered a number of my questions about the SLO wine ordering process, and convinced me to try the (archaic) system again. I am tempted to make this my one and only PLCB store, despite the 30+ minute drive, if only to be nearer to its staff.
Eventually, the manager gave me a stock printout for the Wine Preserver, and told me that I could just take that to my local (East Liberty/Penn Circle) store, I could give it to them, and they would order some from the manufacturer, or from another PLCB store, and I could pick it up at my convenience.
Happy, I went back home. On my way home, I stopped off at the East Liberty PLCB store, and gave the printout to the cashier, and told him what the Robinson Towne Center manager had told me. The E. Liberty cashier looked at me like I was speaking a foreign language, and had me explain it again. He looked at the printout, and saw that a store located 15-20 minutes away had some in stock, and he asked me "why don't you drive out there to pick some up?" After explaining that I wasn't keen on driving out to the THIRD PLCB store of the day, he told me that I could pay 5-10$ to get it shipped to the local store. I agreed to it.
He had me write down my credit card information on the back of a piece of paper. I couldn't believe it was possible, but it was true: the PLCB really does still live in the early 1960's, long before credit card fraud, credit card swipers, and even computers had become common knowledge. Ahhh. Those simpler times. I left without even knowing the total that the PLCB was going to charge to my credit card, and without knowing when I could expect to hear anything from them. Whoever thought that this was a reasonable business practice in the freaking 21st century needs to be drawn and quartered.
Even worse, I learned yesterday that the PLCB was going to stop selling all of its wine-related accessories, including Wine Preserver. Why? Apparently something to do with a "bunch of suits" in Harrisburg thinking that "they weren't making money fast enough." Apparently, individual store managers are even forbidden to carry wine-accessories even if they think it will be profitable or beneficial for their customers. This is a prime example of the near-negligent bureaucratic failures typical of the PLCB.
This story illustrates a few major beefs I have with the PLCB system. First, is that while some employees are surely quite spectacular, somehow, the system apparently lacks any quality control to get rid of or retrain employees that give PLCB a bad name. Second is that the PLCB exhibits the classical monopolistic drawback of detached disservice to its consumers. Why bother improving service if customers cannot go anywhere else, and more importantly, if there is no quality control to ensure that stellar employees are rewarded accordingly?
I stopped by since I was in the area anyway, and I needed some Wine Preserver. Wine Preserver is an ingenious little bottle of nitrogen and noble gases that one sprays into opened wine bottles to keep them from oxidizing and going bad. It works a lot better than those vacuum stoppers.
This trip was particularly notable since it gave me my very first pleasant experience at a PLCB store. The staff were extraordinarily helpful, friendly, and knowledgeable. Even though the store didn't have any Wine Preserver in stock, the store manager (I believe) chatted with me for about 10 minutes as he helped me figure out where to get some. He answered a number of my questions about the SLO wine ordering process, and convinced me to try the (archaic) system again. I am tempted to make this my one and only PLCB store, despite the 30+ minute drive, if only to be nearer to its staff.
Eventually, the manager gave me a stock printout for the Wine Preserver, and told me that I could just take that to my local (East Liberty/Penn Circle) store, I could give it to them, and they would order some from the manufacturer, or from another PLCB store, and I could pick it up at my convenience.
Happy, I went back home. On my way home, I stopped off at the East Liberty PLCB store, and gave the printout to the cashier, and told him what the Robinson Towne Center manager had told me. The E. Liberty cashier looked at me like I was speaking a foreign language, and had me explain it again. He looked at the printout, and saw that a store located 15-20 minutes away had some in stock, and he asked me "why don't you drive out there to pick some up?" After explaining that I wasn't keen on driving out to the THIRD PLCB store of the day, he told me that I could pay 5-10$ to get it shipped to the local store. I agreed to it.
He had me write down my credit card information on the back of a piece of paper. I couldn't believe it was possible, but it was true: the PLCB really does still live in the early 1960's, long before credit card fraud, credit card swipers, and even computers had become common knowledge. Ahhh. Those simpler times. I left without even knowing the total that the PLCB was going to charge to my credit card, and without knowing when I could expect to hear anything from them. Whoever thought that this was a reasonable business practice in the freaking 21st century needs to be drawn and quartered.
Even worse, I learned yesterday that the PLCB was going to stop selling all of its wine-related accessories, including Wine Preserver. Why? Apparently something to do with a "bunch of suits" in Harrisburg thinking that "they weren't making money fast enough." Apparently, individual store managers are even forbidden to carry wine-accessories even if they think it will be profitable or beneficial for their customers. This is a prime example of the near-negligent bureaucratic failures typical of the PLCB.
This story illustrates a few major beefs I have with the PLCB system. First, is that while some employees are surely quite spectacular, somehow, the system apparently lacks any quality control to get rid of or retrain employees that give PLCB a bad name. Second is that the PLCB exhibits the classical monopolistic drawback of detached disservice to its consumers. Why bother improving service if customers cannot go anywhere else, and more importantly, if there is no quality control to ensure that stellar employees are rewarded accordingly?
Tuesday, May 22, 2007
Standard Deviations of the Mean
In my day-to-day life, I run experiments on Database Management Systems, trying to understand their performance characteristics.
Recently, I've been trying to create graphs of average query response times with "standard deviations" --- to ascribe some level of confidence to my data. Unfortunately, I've been doing it all wrong. Fortunately, I have ample company, many people don't quite get it, even in the medical community.
This is the situation: You run an experiment, with thousands of data points (x[i]'s). You want to average them together, and ascribe a confidence to your average.
My first (and incorrect) idea was to compute the mean and the standard deviation from my data points. In other words:
The problem here is that the stdev is an estimate for the variation within the population, and does not have anything to do with the accuracy of your estimate of the mean (mu). In other words, while you expect that your estimate of the mean should get better as you increase the number of data-points, your standard deviations will never get smaller, no matter how many samples you take.
What you really want is Standard Error of the Mean (aka SE or SEM). This is simply the sample standard deviation (above) divided by the square root of the number of samples:
Now you may use the SEM to provide the confidence interval for the mean of your x[i]'s. That is, a 68% likelihood that the true mean is within 1 SEM away from mu, a 95% likelihood that it is within 2 SEM's of mu, and so forth.
So in summary, there are two things of importance to you --- The Standard Deviation of your sample, and the Standard Error of the Mean --- these are very different beasts and should never be confused, no matter how tempted you are to do so.
Recently, I've been trying to create graphs of average query response times with "standard deviations" --- to ascribe some level of confidence to my data. Unfortunately, I've been doing it all wrong. Fortunately, I have ample company, many people don't quite get it, even in the medical community.
This is the situation: You run an experiment, with thousands of data points (x[i]'s). You want to average them together, and ascribe a confidence to your average.
My first (and incorrect) idea was to compute the mean and the standard deviation from my data points. In other words:
mu := sum(x[i])/N
var := sum( (x[i]-mu)^2 )/N
stdev := sqrt(var)
The problem here is that the stdev is an estimate for the variation within the population, and does not have anything to do with the accuracy of your estimate of the mean (mu). In other words, while you expect that your estimate of the mean should get better as you increase the number of data-points, your standard deviations will never get smaller, no matter how many samples you take.
What you really want is Standard Error of the Mean (aka SE or SEM). This is simply the sample standard deviation (above) divided by the square root of the number of samples:
sem := stdev/sqrt(N)
Now you may use the SEM to provide the confidence interval for the mean of your x[i]'s. That is, a 68% likelihood that the true mean is within 1 SEM away from mu, a 95% likelihood that it is within 2 SEM's of mu, and so forth.
So in summary, there are two things of importance to you --- The Standard Deviation of your sample, and the Standard Error of the Mean --- these are very different beasts and should never be confused, no matter how tempted you are to do so.
Thursday, February 22, 2007
Welcome to my Châteauneufs
Eric Asimov, wine writer for the New York Times, has done several recent stories revolving thematically around Chateauneuf-du-Pape wines.
Most interestingly, the "Best Value" Eric's tasting group, the Domaine de Monpertuis 2004, is miraculously available within the PLCB system, and also, at a store near me. The wine is expensive, the PLCB offers it for $30.49, although Eric bought it in NYC for $32. Score one for the bulk-purchasing power of the PLCB. Oh. But wait. You can buy it for $27.99 in Marlton, NJ.
I wouldn't normally want to buy a $30 bottle of wine, but it's a rare occasion that a wine I read about is available to me, and it just so happened that Corina just got a new job. So it was both cause for celebration and cause for wine exploration. When I got to the wine store, I found the wine not on the standard shelves, but on the racks in the special, temperature-controlled back room of the store. I give rare praise --- this climate controlled wine transport and storage is one way that the PLCB does not suck. We have the work of former chairman Jonathan H Newman to thank for that.
So with wine in hand, Corina and I went in for an evening of roasted beets and risotto cakes, and some Châteauneuf-du-Pape. Eric (et al) describe the wine as "earthy," "old-fashioned," and with a balance between "raspberry and tart herbal aromas." Corina and I agreed that the wine was quite good, which makes me proud, since Eric says the wine is "a wine lovers' wine." Our tasting notes give it a dark and earthy raspberry nose and flavor, with an attack that starts sharp, but then lingers mellowly for quite some time. There was something that eluded me in the tasting and the aroma that I could not place. My best guess was mint, but I am doubtful. With all that said, I know how pompously this all reads, and I commiserate. I would buy this again if it cost less than $15, but I'm an explorer, and it's rare I would ever buy any wine again (The exception being the wonderful Kim Crawford Marlborough Sauvignon Blanc, which I have to get).
Most interestingly, the "Best Value" Eric's tasting group, the Domaine de Monpertuis 2004, is miraculously available within the PLCB system, and also, at a store near me. The wine is expensive, the PLCB offers it for $30.49, although Eric bought it in NYC for $32. Score one for the bulk-purchasing power of the PLCB. Oh. But wait. You can buy it for $27.99 in Marlton, NJ.
I wouldn't normally want to buy a $30 bottle of wine, but it's a rare occasion that a wine I read about is available to me, and it just so happened that Corina just got a new job. So it was both cause for celebration and cause for wine exploration. When I got to the wine store, I found the wine not on the standard shelves, but on the racks in the special, temperature-controlled back room of the store. I give rare praise --- this climate controlled wine transport and storage is one way that the PLCB does not suck. We have the work of former chairman Jonathan H Newman to thank for that. So with wine in hand, Corina and I went in for an evening of roasted beets and risotto cakes, and some Châteauneuf-du-Pape. Eric (et al) describe the wine as "earthy," "old-fashioned," and with a balance between "raspberry and tart herbal aromas." Corina and I agreed that the wine was quite good, which makes me proud, since Eric says the wine is "a wine lovers' wine." Our tasting notes give it a dark and earthy raspberry nose and flavor, with an attack that starts sharp, but then lingers mellowly for quite some time. There was something that eluded me in the tasting and the aroma that I could not place. My best guess was mint, but I am doubtful. With all that said, I know how pompously this all reads, and I commiserate. I would buy this again if it cost less than $15, but I'm an explorer, and it's rare I would ever buy any wine again (The exception being the wonderful Kim Crawford Marlborough Sauvignon Blanc, which I have to get).
Wednesday, February 21, 2007
Liquor Control Bites
Until I went to Paris in 2004, I spent my entire life dry. On my 2004 trip to Paris, the heart of the oenophilic, I decided to try as hard as I could to enjoy wine. Every night, I suffered, buying a glass of wine with dinner, and forcing it down, despite my shocking distaste for the swill. I even suffered one awkward night, getting tipsy after just one glass, trying to get back to my hotel without falling in the street. Then it happened: at the end of my 10-day trip, I had a wine that wasn't completely awful. I didn't like it, but it wasn't as bad as the others. I could understand that somebody could like the stuff.
It took about 2-3 more years before I really tried wine again. When I lived in California, I bought cheap wine at Trader Joe's and Whole Foods Sadly, I disliked every wine I bought, but at least I was making an effort (In hindsight perhaps I was just buying too much of the cheap swill). Now that I'm back in Pennsylvania, I help to consume about 2-3 wines per week. Maybe it's just because I'm buying more expensive bottles, but I even like the stuff now.
One of the things I don't like about wine, however, is buying it. More importantly, I hate buying at the Pennsylvania Liquor Control Board (PLCB) stores. By law, they're the only places to buy wine in Pennsylvania. And surprise of surprises, even though it's a government-run monopoly, the PLCB sucks.
Going to a PLCB store is a singular experience, much like going to a concert performed entirely by the deaf. Most of the employees I've encountered have no understanding of the product which they sell. Since there is no competition, it doesn't matter if the employees know anything at all, they just need some alcohol for sale. The best wine information I've ever gotten from a PLCB employee is "This is on sale for under 8$! It must be great!" and "I don't know what it tastes like, but it always sells out, so it must be good!"
Also due to the lack of competition, consumer choice is entirely at the mercy of the PLCB. The poor selection is what infuriates me most as a consumer --- when you want something particular, you're screwed. After reading a wine magazine, newspaper article, or blog, and I want to try a "hot new wine," inevitably I cannot buy it. Most of the time, say around 80%, the PLCB simply hasn't even heard of the wines or vintners. About 15% of the time, the PLCB has heard of it, but either has never purchased any, or has no more left. Of course, they cannot order these for you (except in very special circumstances, and when you want to buy a few cases). 5% of the time, they have some in stock, but not at any store within 200 miles. Of course, they cannot ship some stock to a store near me. The remaining 1% of the time, I can get my wine. ("Fortunately," the PLCB web site allows you to search for product availability. I should warn you, however, that that it will be far easier to pound an 8 inch nail through your skull than to use that site for any length of time).
Thus, I have believed the PLCB was The Devil, existing only to hurt Pennsylvanians, and that the only way to Happiness and Morality was the demise of the PLCB. This model of the world, however, was shattered on a recent trip to Cleveland. I was buying a wine, and found the bottle was 3-4$ more expensive than in the PLCB. I couldn't believe it: it was better to buy that wine in Pennsylvania than in Cleveland.
I figured that the Pennsylvania and adjacent non-control wine markets must have been extensively compared online, and I could learn whether this was a trend or not. All I found online, however, was a testimonial from Vice Carocci discussing historical changes to the PLCB. Vince argues that the PLCB has gotten much better, and today helps Pennsylvanians get better prices. I wanted to examine this hypothesis more carefully.
The PLCB is the second-largest single buyer of wine and liquor in the world (after Ontario's liquor control board). As such, it may indeed have the power to negotiate unbelievably low prices. In fact, PLCB stores have drunk of Vince's Kool-Aid, and have signs touting these extremely low prices. Sadly, all the PLCB's information is misinformation. They compare their retail price to vintners' "suggested prices," which are far higher than anybody pays for the wine. What they should do is compare the price Pennsylvanians pay for wine to the price Ohians and Jersyites pay for the same wine.
That is the comparison I hoped to make. Using the online wine store WineZap and the PLCB Web Site, I did a quick survey of wine prices, and came up with the following sample data:
| Wine | Internet $$ | PLCB $$ | Ratio |
|---|---|---|---|
| Brundlmayer Riesling Steinmassel 2004 | 25.99-30.00 | 21.49 | 0.82 |
| P Jaboulet Aine Cotes-du-Rhone | 17.99 | 18.79 | 1.04 |
| Belle Glos Pinot Noir Taylor.Ln 2004 | 46.40 | 49.99 | 1.07 |
| Annies Lane Riesling 2005 | 12.00 | 12.99 | 1.08 |
| Carretta Barbaresco Bordino 2000 | 35.99 | 39.49 | 1.09 |
| Chateau Mont Redon Cotes-du-Rhone 2003 | 14.35-15.99 | 16.59 | 1.15 |
| Kim Crawford Marl. Sauv Blanc | 12.99-24.99 | 14.99 | 1.15 |
| Ch Doisy Vedrines Sauternes 2001 | 36.99-69.00 | 42.99 | 1.16 |
| Ch Piada Sauternes 2001 | 62.99-69.99 | 74.59 | 1.18 |
| Ch St Michelle Eroica Riesling 2004 | 17.99-26.29 | 21.99 | 1.22 |
I chose the wines more or less haphazardly from some wine varieties that I would like to try, and in several different price brackets.
The quick summary is as follows: Only in one case did the PLCB have the best price available. In about half the wines, the PLCB was competitive --- within 10% of the best price available, but the other half, the prices were significantly higher. If you have to buy your wine, buy it when you're not in Pennsylvania. You'll get a better selection and (if you look for it) a better price. Of course, almost NONE of the wines listed above could actually be purchased at a PLCB store near me, even though I could buy those wines online if I lived in Ohio.
More detailed analysis, however, will indicate that the PLCB really does a good job of securing good prices for its wines. The problem is Pennsylvania alcohol taxes. The Johnstown Flood Tax imposes an 18% tax on all wine (contained in the PLCB prices above). If you remove that 18% tax from the PLCB prices above, the PLCB has the best price for all but one of the wines above.
This is already a long post, and there is much to say, so I will return to these subjects in future posts. In the meantime, I want to tell you that the PLCB is probably still your enemy, but at least it could give you good prices, if it weren't for state taxes.
