Simon blogs

[namida]

Nothing is free, the cashback customers are paying with their time and lost privacy.

I would question this in two ways - firstly, whether the sale will really be private in this day and age even without a loyalty / cashback card; and secondly, whether it's really that big a deal that some giant marketing firm who's never heard of you beyond as "customer #410757864530" knows which brand of toothpaste you buy.

I do however see the point in regards to it slowing down transactions. Stores here often handle this by allowing the customer to swipe their loyalty card while the cashier is ringing up their items, thus, no additional time is lost. Also, in my experience here from working in said stores in the past (which may not be the same in Europe), the majority of customers do tend to have these cards, although it's questionable how many ever bother to get the rewards (most customers are likely more interested in the additional, on-the-spot discounts - which can actually be quite significant, sometimes as much as $15 - $20 off a $100 order).

[nin10doadict]

Sometimes I go shopping with coupons in my pocket, forget they're there, and don't use them. The savings are nice but really not enough to merit me seeking them out; I just try to be frugal in other ways, such as keeping track of the prices of certain items and only buying them when the price drops. There are some cases where this is useful, because these items will always be marked as 'on sale' even when the price is high, so mentally tracking the price helps me determine when the so-called sale is actually real.

[mobius]

if you go out of your way to get or use a coupon (buying something you wouldn't have bought otherwise) then you're not saving anything. On the other hand if you use one that happens to fall into your lap (as sometimes happens) on something you were planning on buying anyway; it does save money, though a small amount. Over time it adds up if you're diligent with this method.

[ccexplore]

I sort of get what Simon's saying, but maybe it's a little disingenuous to pick the worst case example where the coupon isn't working; far fewer people would bother with coupons if they actually frequently break down like your example.  It'll be kind of like applying a worst case example where the computer system has bad price information so barcode scanning an item simply on sale (ie. no coupon involved) did not result in the correct price, and then the junior clerk has to go get a more experienced clerk to sort out the mess.  You can't exactly use that example to say the barcode scanning system is bad, unless it happens frequently.

I also always thought coupons are more the store's sneaky way to try to have their cake and eat it too.  If they just simply put the item on sale for everyone, then everyone pays the same lower price (but yes, maybe it'll attract more people to buy in their store as opposed to some other place).  But if they force you to cut out a coupon in order to get the lower price, at least some portion of people might not bother but still want to buy, and can then pay the higher price, while the truly price-conscious/obsessed people will take the extra time and feel good about getting the discount, and the store at least gets them to buy here as opposed to some other place.  Yes, you lose a few people who'd reject both the higher price and the work needed to cut out the coupon, but maybe despite that, it still balances out in favor for the store?

[Minim]

This is right up my alley as I work as a cashier in my local supermarket. I feel your pain too as most of my regular customers suffer similar agonising waits as well.

Sometimes I have to ask a young-looking customer for their ID for any purchase of cigarettes or alcohol. 18 is the golden age for this, but annoyingly I have to follow the same rule for 16 year-olds buying Red Bull cans. On another day I sometimes need to ask for the same customer's ID presentation. It can infuriate them, particularly if they know me so well, so I only do this if managers are walking around, just in case they wag their finger at me.

Most customers buy fruit, vegetables and fresh bread without sticking a barcode sticker on them. The hard work has to be done by me to find the correct item on the system. The breads are most difficult because the pictures look so similar, especially the sourdoughs; And I can never tell a loaf from a crusty bread either. Anyway, afterwards, I sometimes have to weigh the product instead of add the quantity. (Sometimes the check weigh doesn't work, which consumes even more time)

As well as this, the system tells me to ask each customer for:

• Their loyalty card. The main advantage for our shop is a free newspaper for £10 spent, including the paper. Some of our customers do, but some of them present me coupons before they show their loyalty card. If I try put coupons through the system before the loyalty card I get in trouble and need to ask a manager.
• Cashback. This only comes up when customers place their debit card into the machine, instead of the simple contactless process. Some of our customers are fussy about how they would like their cashback, e.g. in coins.
• Swipe their card: Some customers' cards don't work on our card machines. Sometimes this can be circumvented by me holding the card in, but after three failed attempts I swipe their card if there is a signature on the back of it. If there isn't then I have to ask the customer to pay with another card, or cash. (Sometime I ask them to use the cashpoint if they don't have cash, which fortunately isn't too fay away, but it's still an unnecessary waste of their time)

At the end of each transaction the receipt comes out, but occasionally another coupon comes out. Rather than feel guilty about not giving the customer the extra offer I always wait a few more seconds after the receipt prints off just to see if the additional offer will come out as well.

And that's all I can think of for now. Hope it's not too off-topic but it certainly feels like we're damaging our reputation with all these delays for these poor customers.

[ccexplore]

Interesting that in the UK you do barcodes for fresh produce.  In the US I've never seen this so far, for fresh produce you (or more likely, the cashier) always have to do manual entry.  On the other hand for things like meats and seafood, the butcher/fishmonger does apply the label with barcode as part of the process of weighing, cutting and packaging of the meat/seafood, so all you or the cashier has to do afterwards at checkout is to scan it.

In the US at least for the large corporate supermarkets, it seems there usually is a separate place in the store where you buy alcohols and cigarettes, you wouldn't even be dealing with those items in the regular checkout lines.  [edit: actually now I think of it, that only applies to cigarettes and other tobacco products, not alcohol.  I guess cashier would ask for ID on alcohol purchases.  Kind of curious now what would happen with alcohol in the self checkout lanes.]  Of course I imagine for smaller supermarkets the same cashier would handle regular and restricted items.  Why does Red Bull require IDs in the UK?  It isn't alcoholic AFAIK.  In the US there is no legal regulation on caffeine AFAIK (certainly no federal regulations; state and local laws can in theory do this but I've never heard of it).

[Simon]

firstly, whether the sale will really be private in this day and age

Mostly yes, I pay cash for 90 % of my purchases.

Sweden: Customary to take bank cards for any bill, and people happily pay everything with that. Germany: Most places don't accept it unless the bill is 5 Euros. Thus cash is still popular.

Online purchases are a pain. The super-high-secure checkout sends unencrypted email afterwards.

whether it's really that big a deal that some giant marketing firm who's never heard of you beyond as "customer #410757864530" knows which brand of toothpaste you buy.

I thought cashback programs had you register with name? At least banks do. They'll keep this data in their super-high-secure datacenter, which falls prey to social engineering 3 years down the road. Or to an overworked maintenance programmer.

The main problem is that other people use cashback. That is the real time drain.

At my store, the cashback ad says:
Exchange your cashback points for vouchers!
200 points = voucher worth 2 Euros
500 points = voucher worth 5 Euros
1000 points = voucher worth 10 Euros
2000 points = voucher worth 20 Euros

Makes one wonder at how low a social level this list is aimed.

Also, calling the cashback point a "point" instead of 0.01 Euros. Who would swipe their card for 3 cents? Being unbothered must be worth more to even those people. But 3 points, sure, that's worth it! Numbers go up! Fun!

items will always be marked as 'on sale' even when the price is high

Hah, such simple technique, and I'm sure it works more often than it should. I haven't seen it conciously, but still sound advice.

(most customers are likely more interested in the additional, on-the-spot discounts - which can actually be quite significant, sometimes as much as $15 - $20 off a $100 order).

Yes, I agree that this is a lot, and I would even let myself be bothered to get an extra 10 to 20 Euros. Saving 10 % is nice.

on something you were planning on buying anyway; it does save money, though a small amount. Over time it adds up if you're diligent with this method.

I accept this. I'd say it costs too much time for the effort, and you risk end costing other people's time. If your experience is less annoying than mine, that's fine.

it's a little disingenuous to pick the worst case example where the coupon isn't working;

It may be slightly over-the-top, but it's still representative, sadly. There are many other ways that produce excessive waits:

• Old lady must be told several times that now is voucher/cashback time, then she digs in wallet for cashback card.
  
• Person (even smart ones) forget to put vouchers on the counter until after the system seemingly-randomly expects vouchers to be entered. Transaction rollback is problematic, sometimes needing special keys or privileged members. (Also see Minim's post.)
  
• Person wants to mix two types of voucher/cashback. How does it go? Ask the cashier risk other people's time? Memories of clashing features in computer programs come up. Even more memories of house rules in games come up that hook into the same event of the basic game and their interaction/timing is ill-defined.
  
• The cashpack program has an annoying mascot that looks like on drugs, and it stares into your eye everywhere in the supermarket. By checkout time, I want to bite somebody's head off. I accept that this maybe doesn't apply to everybody. >_>;;
  

Every voucher/cashback program is one extra way how designers, over the years, manage to screw up every simple idea (buying at a physical store). The nastiness of these programs grows at O(n^2).

You can't exactly use that example to say the barcode scanning system is bad, unless it happens frequently.

Barcode scanning is one alternative out of many mutually exclusive ways to register prices. It is far less error-prone. It's also ubiquitous and has grown bug-free. It doesn't collide with other cash-out-time stuff.

Coupons fail far more often. Certainly I accept that my specific failure is rare, but they can fail in many more ways, even if less time-consuming.

Some of our customers do, but some of them present me coupons before they show their loyalty card. If I try put coupons through the system before the loyalty card I get in trouble and need to ask a manager.

This is beautiful.

I feel like all clerks should fail this step on purpose, every single time this ordering matters. The manager should be called every time. That is how nasty it is.

This really should have come up in user testing: Supermarket-manufacturing firms probably test their checkout processes on real clerks, providing fake customers with random such problematic cases.

The many kinds of different vouchers/cashback can't possibly be known at design time (when the supermarket-cashier-lane-manufacturer designs his cashier user interface), and no single voucher/cashback firm should be responsible to define all interactions with all other voucher/cashback firms. This is the main reason why vouchers/cashback are fundamentally flawed from a systems design point-of-view. Nobody can be made responsible for the mess. My hunch is then to cut such mess immediately from the design. (Of course my hunch won't make the marketers any money.)

Why does Red Bull require IDs in the UK?

At least in Germany, there is only one age check for alcohol and tobacco at age 16. Hard liquor is locked behind glass doors and customers must ask a clerk to unlock.

Interesting that the UK introduces so many more more special rules.

-- Simon

[namida]

Mostly yes, I pay cash for 90 % of my purchases.

Sweden: Customary to take bank cards for any bill, and people happily pay everything with that. Germany: Most places don't accept it unless the bill is 5 Euros. Thus cash is still popular.

Interesting. It's quite different here; while credit cards sometimes do have a minimum spend (often in the region of $10 or so), debit cards generally do not, and many shops encourage customers to use debit cards rather than cash. Even without that encouragement, the general trend seems to be to prefer card here. I worked in supermarkets for a while in the past, and while cash was far from unheard of, card was definitely the more popular option - and I've heard this has only become even more the case since then.

It probably helps that fees are very minimal here. If you're happy to settle for a card that can only be used in NZ (or at certain international ATMs for a fee), it is possible to literally pay no fees - none on transactions, no monthly/yearly fees, I literally mean nothing at all. If you need online / international use of your card, you'll need a higher-tier card, but this still only costs in the region of $5 per year - and you can still avoid all the other fees.

I thought cashback programs had you register with name? At least banks do. They'll keep this data in their super-high-secure datacenter, which falls prey to social engineering 3 years down the road. Or to an overworked maintenance programmer.

Sure, but to them, "John Smith" is as meaningless as "customer #410757864530". They don't know who John Smith is, as such, just that a guy called John Smith (maybe "who lives at 123 Fake Street") buys lots of chocolate, or whatever. The worst they can do with this is send aforementioned John Smith advertisements for more chocolate. A minor annoyance, but I guess I kinda got over seeing it as being a huge deal. I guess after dealing with internet spam about "enlarge your penis!" and "buy [obscure prescription drug] cheap!", I'm not so bothered by offline ads that are at least somewhat relevant.

Now of course, there may be cases where anonymity is preferable. If you're buying, say, ingredients / equipment to grow / produce drugs, you probably don't want a record of that. But if you're simply buying bread, pasta and toothpaste, I really don't see how this is a big deal anymore.

[ccexplore]

One more thing I want to add, in the same vain as my earlier point how all this probably benefits the store more than the customers:

Yes, those things you mentioned waste everyone's time, but I think the store has much more capacity to absorb the extra times spent compared to individual customers, even when it comes to major issues like the example you cited.  Because the store is pretty much never equally busy at all hours of day, extra time spent here and there will usually be balanced out by lulls elsewhere during the day.  Also, it's probably comparatively rare that a customer who's already in line ready to pay will decide to give up the purchases altogether due to issues with someone in front of the line; in almost all cases they'll simply wait the extra time, very annoyed for sure, but ultimately the store does not lose the transaction from the customer.

Now yes, on a longer term basis, if a store consistently and frequently runs into these kinds of issues, some customers will definitely stop going to the store in preference for another place without such frequent issues, so it is still in the store's interest to minimize these kinds of issues.  But beyond a certain threshold of "it doesn't happen often enough, or take up too much time or otherwise impact customers negatively", the store basically can absorb the occasional extra major time spent (and the more frequent little extra seconds here and there) with minimal risk of loss purchases or loss customer retention.  And so there's no incentive from the store's perspective to get rid of the coupons or loyal programs or whatever, despite the inefficiencies they can introduce to the checkout process.

To put it differently:  yes, what happened sucked for Simon.  But will that rather unfortunate incident be enough to get even Simon to permanently boycott the store and shop elsewhere?  I'm guessing no?

[Proxima]

We should perhaps also consider the collective time cost of the incident causing everyone to make all these posts about it

[Simon]

Right, I wouldn't boycott the store because of this one incident. Even though the store and the cashback program make money from this at the expense of customers' time, it's better for me to stay with this store than to walk anywhere further.

I enjoy the time spent ranting. It may have been longer than the wait in the shop, but rants free the mind.





Let's all be merry and cheerful and in holiday mood! Cookies and tea for everybody's enjoyment. Cinnamon odor in the air. Red and golden ornaments. And animated tree gifs for everybody's homely nostalgic 90's website feelings.

How shall we be merry today? No focused rant about a single topic, but some small thoughts instead.

There is no widespread good Windows package manager. Most widespread is Steam, but that's closed-source malware and hard to get your software accepted. NuGet is promising, but who has heard of it? I bet that, were I to ship Lix with NuGet, everybody on the forum would still download Lix manually. You wouldn't want to install a package manager for one single application.

Chess world championship, they drew 12 games out of 12. I've only watched the 12th game on livestream. I didn't even pay attention to the scheduling; instead, I was hacking on my Lix release scripts to auto-build Win32, Win64, and Linux 64-bit. Still, the game commentary was exciting enough that I ditched my planned nap and drank 4 liters of tea instead.

Shogi (Japanese chess with flowery-named pieces such as odor cart) looks like a nice game, favoring romantic attacks even more. No more woodpushing to draw 12 out of 12, instead happy tactical king hunting every time!



Of course one would have to learn the kanji first, but let us assume that everybody can read the kanji easily. Even then, Shogi has massive UI problems.

Pieces should be colored by type. Piece ownership is indicated by pointing direction, but for this, the pieces should look much more triangular. Their existing obstuse arrow shape is almost rectangular. Some people have invented pieces that you can more easily recognize.



A possible approach, but these should be colored by type, and they should be much pointier.

But now, the roadblock is that this would never last beyond practice games. Even westerners easily memorize the kanji after a few games. Any tournaments would be played with the traditional pieces anyway. If one were to learn Shogi, one should use kanji pieces as early as possible, to reduce friction.



Compare the Shogi situation with Skat, a popular card game in Germany that is also played on a high level at tournaments. Here, the 4-color deck ("Turnierbild") has become tournament standard. It shows that some excellent UI improvements can overcome the hump of other people's accustomization to inferior standards.

40 years ago, in West Germany, 2-color French decks were standard, and in East Germany, the German deck with 4 completely different suits in 4 colors was standard. I believe that the Turnierbild has only gained enough traction because there was a need for a universal deck, and the Eastern Germans were already accustomed to 4-color decks. But anyway, the standard improved across Germany, which is what counts in the end.

Sadly, Turnierbild isn't used in Doppelkopf tournaments. At least my peer group at university played Doppelkopf exclusively with the 4-color Turnierbild.

In online gaming, you can choose your deck/pieces independently from what the opponents see on their machines. That's helpful to give less popular standards an opportunity to rise. All my UI considerations today are about offline games where everybody sees the same standardized playing material.

Ah, the some small thoughts became a UI rant again.

-- Simon

[nin10doadict]

You gave up your nap? Clearly you don't have enough pets. I have discovered that keeping the heat turned down low not only saves on the heating bill but encourages the pets to come and nap with me. Had an hour long nap yesterday because I had lost sleep Sunday night. Big Nubs actually crawled under the blankets with me, which he hardly ever does. Getting up from that nap was one of the hardest things I've ever had to do.

12/12 games drawn? I wonder how many of the 'draw game' rules they obey in the chess tournament. I would assume all of them, even the obscure ones like 'game is drawn if 50 turns pass with no captures or pawn movement' or 'game is drawn if the layout of the pieces is the exact same three separate times.' Keeping track of those without some sort of computer would be rather difficult, I think.

[Proxima]

12/12 games drawn? I wonder how many of the 'draw game' rules they obey in the chess tournament. I would assume all of them, even the obscure ones like 'game is drawn if 50 turns pass with no captures or pawn movement' or 'game is drawn if the layout of the pieces is the exact same three separate times.' Keeping track of those without some sort of computer would be rather difficult, I think.

The rules are actually: a draw may be claimed if 50 turns pass with no captures or pawn moves, or if the same position repeats three times. It's not automatic. Both are very rare in actual play; nearly all draws are by agreement. But these rules are part of the laws of chess, so they always apply.

Serious players are required to write down their moves, so it's easy to see when a 50-move situation is coming up, and for threefold repetition, the most likely situation is moving back and forth, not repeating a position from earlier in the game; captures would tend to prevent that.

(Although hilariously, there was a game where nothing was captured, and both sides' pawns got into a deadlock that prevented any progress. One player was stubborn and wouldn't agree a draw, and was confounded when his opponent successfully claimed one under the 50-move rule. I don't know who the players were; I found this game on Tim Krabbé's website that collects unusual games.)

[Dullstar]

I have mixed feelings about package managers. I've experimented with Linux in the past (though not so much recently), and one of the common frustrations I had was that the packages in the repositories were frequently out of date. But it's convenient for browsing for available software, and also convenient for automatic updates. One of my constant frustrations with NeoLemmix and Lix is updates.

[ccexplore]

Shogi (Japanese chess with flowery-named pieces such as odor cart)

"odor cart":  sounds like a bad translation perhaps?  I believe with languages like Japanese if you attempt to only translate individual Kanji characters in isolation and ignoring contexts, you can sometimes end up with results that have almost nothing to do with the actual words or phrase in question.  Also, doesn't some Kanji characters have multiple very different meanings (almost like homophones but actually written exactly the same way as well?).  Do we know if "odor cart" isn't just creative cherry-picking on a particular reading of the Kanji in question, even though it isn't actually the intended/correct reading being used wrt the name of that particular Shogi piece?

Pieces should be colored by type. Piece ownership is indicated by pointing direction, but for this, the pieces should look much more triangular. Their existing obstuse arrow shape is almost rectangular. Some people have invented pieces that you can more easily recognize.

I went to wikipedia article for Shogi, and within the first sentence where it also mentions other chess-like variants in Asia with links to respective articles, I clicked through all of them and they all do use different colors to denote sides.  Japanese Shogi seems unique in not using colors to differentiate sides.

That being said, if you can actually read Kanji, I think perhaps it's sufficient to tell that some piece's Kanji are upside down from your POV, and therefore not your side?  I do agree though, a more prominent visual way to denote sides would seem preferable, especially for spectators.

On the other hand, none of the variants used different colors for different types (if by "type" you mean like pawns vs kings or whatever their equivalents are in Shogi).  Then again, neither does Western chess, although in that case the types differ by shape.  It seems for the Asian variants listed in Wikipedia, they either use different shapes for types, or else I guess the difference in Kanji is sufficient for the variants where Kanji are used on the pieces to denote types.