In other news today...


  • Discourse touched me in a no-no place

    @loopback0 said in In other news today...:

    I had most of my Uni lessons across the road from there (the John Dalton building). It was long enough ago that the BBC was still across the road though, so a long time before that place sprung up.

    The (old) BBC site is now a huge construction site (as is a lot of the rest of the area). At the moment, they've got the main lift shafts done — a mixture of 15-storey and 35-storey structures — and are working on the steel frame for the rest of the building.


  • 🚽 Regular

    First the FAA abdicating from its responsibility and now the FDA.


  • BINNED

    @anonymous234 said in In other news today...:

    @DogsB Having to use human cashiers will reduce the "generational divide"? :wtf: This does not even begin to make sense.

    Is it because young and old people tend to use different options, so we must shame the young people into rejecting their preferred choice so older people won't have to see others with slightly different behavior to their own?

    Yeah, I hate it when young people use credit cards to pay their pack of gums, and it takes forever. OTOH, I also hate it when old people... well, they take forever anyways no matter what they do.
    So, really, the problem is people. Give me a supermarket without that, TYVM.

    Filed under: or at least install a shot-clock and make everyone hurry TF up.


  • 🚽 Regular

    @topspin said in In other news today...:

    @anonymous234 said in In other news today...:

    @DogsB Having to use human cashiers will reduce the "generational divide"? :wtf: This does not even begin to make sense.

    Is it because young and old people tend to use different options, so we must shame the young people into rejecting their preferred choice so older people won't have to see others with slightly different behavior to their own?

    Yeah, I hate it when young people use credit cards to pay their pack of gums, and it takes forever. OTOH, I also hate it when old people... well, they take forever anyways no matter what they do.
    So, really, the problem is people. Give me a supermarket without that, TYVM.

    Filed under: or at least install a shot-clock and make everyone hurry TF up.

    Do they have tap-to-pay in your region? It's everywhere here now and it seems to work well, I've generally stopped carrying cash as there is no need. I wasn't on board with using my mobile-phone to pay for stuff but NFC credit/debit cards is not a bad idea.


  • BINNED

    @Cursorkeys said in In other news today...:

    Do they have tap-to-pay in your region?

    Yeah, that's catching on. I'm not sure if it's faster yet because that just means people ask first if the machine supports it and fiddle around, but that's a temporary problem.

    (I'm still not a fan of paying small amounts in anything other than cash, but that's for different reasons)



  • @topspin Meh. Paying for a pack of gums with a contactless card is way faster than anything old people do. Especially anything involving old people and coins. (Thank god for designated cashless registers.)

    The thing that slows stuff down is the badly implemented policy for randomly checking the PIN for contactless payments even when they are below the PIN-free limit. It's not a bad idea per se, it's just that the implementation from my bank does this by causing the transaction to fail (presumably from their end?), which results all sorts of timeouts before anybody can do anything, and then requires the cashier to restart the transaction from their end.


  • ♿ (Parody)

    @pie_flavor said in In other news today...:

    @TimeBandit said in In other news today...:

    @pie_flavor won't be happy

    Specifically, they noted that the updated draft language proclaims that consumers would not need to read a contract to be bound by its terms

    The draft states as long as consumers received “reasonable notice” and had “reasonable opportunity to review” it

    which is literally what I had been saying in the first place,

    False. Though it may have been what you had meant to say.


  • ♿ (Parody)

    @topspin said in In other news today...:

    @anonymous234 said in In other news today...:

    @DogsB Having to use human cashiers will reduce the "generational divide"? :wtf: This does not even begin to make sense.

    Is it because young and old people tend to use different options, so we must shame the young people into rejecting their preferred choice so older people won't have to see others with slightly different behavior to their own?

    Yeah, I hate it when young people use credit cards to pay their pack of gums, and it takes forever.

    What takes forever about it? Seems like when I pay for stuff it doesn't take any longer than anyone paying cash and waiting for change. Probably quicker in most cases, at least for small amounts where you don't need to sign.



  • @cvi said in In other news today...:

    it's just that the implementation from my bank does this by causing the transaction to fail

    That seems to be crappy bank/card company. For me the terminal occasionally simply says “contactless payment not possible; insert chip” and inserting the chip continues the transaction normally.


  • Considered Harmful

    @boomzilla said in In other news today...:

    @pie_flavor said in In other news today...:

    @TimeBandit said in In other news today...:

    @pie_flavor won't be happy

    Specifically, they noted that the updated draft language proclaims that consumers would not need to read a contract to be bound by its terms

    The draft states as long as consumers received “reasonable notice” and had “reasonable opportunity to review” it

    which is literally what I had been saying in the first place,

    False. Though it may have been what you had meant to say.

    You're welcome to quote otherwise.


  • BINNED

    @boomzilla said in In other news today...:

    @topspin said in In other news today...:

    @anonymous234 said in In other news today...:

    @DogsB Having to use human cashiers will reduce the "generational divide"? :wtf: This does not even begin to make sense.

    Is it because young and old people tend to use different options, so we must shame the young people into rejecting their preferred choice so older people won't have to see others with slightly different behavior to their own?

    Yeah, I hate it when young people use credit cards to pay their pack of gums, and it takes forever.

    What takes forever about it? Seems like when I pay for stuff it doesn't take any longer than anyone paying cash and waiting for change. Probably quicker in most cases, at least for small amounts where you don't need to sign.

    Depends on how it's done. When it's just swiping without signing or PIN, sure it can be fast.
    It doesn't really matter, the point was more that people take forever, even when they could be faster. The only real solution would be borrowing @dkf's cattle prod.


  • ♿ (Parody)

    @pie_flavor said in In other news today...:

    @boomzilla said in In other news today...:

    @pie_flavor said in In other news today...:

    @TimeBandit said in In other news today...:

    @pie_flavor won't be happy

    Specifically, they noted that the updated draft language proclaims that consumers would not need to read a contract to be bound by its terms

    The draft states as long as consumers received “reasonable notice” and had “reasonable opportunity to review” it

    which is literally what I had been saying in the first place,

    False. Though it may have been what you had meant to say.

    You're welcome to quote otherwise.

    You, too.


  • Java Dev

    @topspin said in In other news today...:

    @anonymous234 said in In other news today...:

    @DogsB Having to use human cashiers will reduce the "generational divide"? :wtf: This does not even begin to make sense.

    Is it because young and old people tend to use different options, so we must shame the young people into rejecting their preferred choice so older people won't have to see others with slightly different behavior to their own?

    Yeah, I hate it when young people use credit cards to pay their pack of gums, and it takes forever. OTOH, I also hate it when old people... well, they take forever anyways no matter what they do.
    So, really, the problem is people. Give me a supermarket without that, TYVM.

    Filed under: or at least install a shot-clock and make everyone hurry TF up.

    I generally pick my line almost exclusively based on the number of people in it. The housewife with a full cart tends to barely take more time than the millennial buying a single microwave lunch.


  • Discourse touched me in a no-no place

    @PleegWat said in In other news today...:

    The housewife with a full cart tends to barely take more time than the millennial buying a single microwave lunch.

    Just wondering whether this is a positive observation about the housewife, or a less positive observation about the millennial...



  • @topspin said in In other news today...:

    @boomzilla said in In other news today...:

    @topspin said in In other news today...:

    @anonymous234 said in In other news today...:

    @DogsB Having to use human cashiers will reduce the "generational divide"? :wtf: This does not even begin to make sense.

    Is it because young and old people tend to use different options, so we must shame the young people into rejecting their preferred choice so older people won't have to see others with slightly different behavior to their own?

    Yeah, I hate it when young people use credit cards to pay their pack of gums, and it takes forever.

    What takes forever about it? Seems like when I pay for stuff it doesn't take any longer than anyone paying cash and waiting for change. Probably quicker in most cases, at least for small amounts where you don't need to sign.

    Depends on how it's done. When it's just swiping without signing or PIN, sure it can be fast.
    It doesn't really matter, the point was more that people take forever, even when they could be faster. The only real solution would be borrowing @dkf's cattle prod.

    I think even with pin it's faster than either counting out exact change, or giving cash and having the cashier count your change.



  • @PJH neither. Transaction time is dominated by the payment, regardless of the amount of goods. Even with the fast methods or people.



  • @Benjamin-Hall That has not been my experience, at least here in Canuckistan. Most people get their payment done pretty quickly, other than the odd person out who has coupons or their card doesn't work or whatever. The biggest part of the time is spent scanning and bagging IME


  • ♿ (Parody)

    @Benjamin-Hall said in In other news today...:

    @PJH neither. Transaction time is dominated by the payment, regardless of the amount of goods. Even with the fast methods or people.

    That is the extreme exception to my experience, which sounds a lot like @hungrier's. I would add that I look out for people using their WIC vouchers because those require a lot of extra work by the cashier (as I said, an exceptional circumstance).


  • BINNED

    @hungrier said in In other news today...:

    @topspin said in In other news today...:

    @boomzilla said in In other news today...:

    @topspin said in In other news today...:

    @anonymous234 said in In other news today...:

    @DogsB Having to use human cashiers will reduce the "generational divide"? :wtf: This does not even begin to make sense.

    Is it because young and old people tend to use different options, so we must shame the young people into rejecting their preferred choice so older people won't have to see others with slightly different behavior to their own?

    Yeah, I hate it when young people use credit cards to pay their pack of gums, and it takes forever.

    What takes forever about it? Seems like when I pay for stuff it doesn't take any longer than anyone paying cash and waiting for change. Probably quicker in most cases, at least for small amounts where you don't need to sign.

    Depends on how it's done. When it's just swiping without signing or PIN, sure it can be fast.
    It doesn't really matter, the point was more that people take forever, even when they could be faster. The only real solution would be borrowing @dkf's cattle prod.

    I think even with pin it's faster than either counting out exact change, or giving cash and having the cashier count your change.

    You need faster cashiers.



  • @topspin said in In other news today...:

    It doesn't really matter, the point was more that people take forever, even when they could be faster.

    You sound like you would love Amazon's store





  • @jinpa said in In other news today...:

    prior to about 10 years ago, I was never pressured into signing that I had read something.

    Obviously, you've never purchased a home!



  • @dcon said in In other news today...:

    @jinpa said in In other news today...:

    prior to about 10 years ago, I was never pressured into signing that I had read something.

    Obviously, you've never purchased a home!

    Actually, I have. The only time I remember ever having to state that I had read something was when I've bought something on-line. As grisly as the settlement process was, I don't think I had to state that I had read something. I did read everything (even though they would have preferred that I hadn't), but I didn't have to sign that I had read it.




  • Java Dev

    @hungrier said in In other news today...:

    @Benjamin-Hall That has not been my experience, at least here in Canuckistan. Most people get their payment done pretty quickly, other than the odd person out who has coupons or their card doesn't work or whatever. The biggest part of the time is spent scanning and bagging IME

    IME, cashiers are quite fast at scanning. Bagging (done by the customer) can be a bottleneck, but checkouts are commonly designed so a new customer's items can be scanned while the previous customer is still bagging.

    I have had occasions where the person in front of me had less groceries than I have, but I was done bagging before they were because I tend to transfer items into my bags as the cashier's scanning them.



  • @PleegWat said in In other news today...:

    @hungrier said in In other news today...:

    @Benjamin-Hall That has not been my experience, at least here in Canuckistan. Most people get their payment done pretty quickly, other than the odd person out who has coupons or their card doesn't work or whatever. The biggest part of the time is spent scanning and bagging IME

    IME, cashiers are quite fast at scanning. Bagging (done by the customer) can be a bottleneck, but checkouts are commonly designed so a new customer's items can be scanned while the previous customer is still bagging.

    I have had occasions where the person in front of me had less groceries than I have, but I was done bagging before they were because I tend to transfer items into my bags as the cashier's scanning them.

    On a slight tangent, I have noticed that in department stores (e.g. Target), the lines move much more slowly than they do in supermarkets, even though the customers typically have far fewer items than they do in grocery stores.


  • I survived the hour long Uno hand

    @jinpa
    It does take quite a while to remove all those ink bomb loss prevention tags from the clothing. And sign the customer up for the three new store brand credit cards and six mailing lists for new deals & coupons.


  • BINNED

    @izzion While we’re going off-tangent, that reminds me of a situation in the supermarket last week. Unfortunately, because of word order this only works in German and not when translated, but I’ll translate it anyway.

    Supermarket radio: “Don’t want any of our coupons anymore...”
    topspin sigh “fuck, no I don’t.”
    [end of the sentence such that the final sentence was]
    Supermarket radio: “[Don’t want] to miss [any of our coupons anymore]?”
    topspin another sigh



  • @topspin I'm speaking German (and English) as you know but I fear that I cannot make heads or tails out of that in either language. ❓



  • The only way this pairing could be more perfect is if anyone was eagerly anticipating the Galaxy Fold in the first place.


  • BINNED

    @Rhywden said in In other news today...:

    @topspin I'm speaking German (and English) as you know but I fear that I cannot make heads or tails out of that in either language. ❓

    “Wollen Sie keine unserer Coupons mehr...”
    🖕
    “... verpassen?”


  • ♿ (Parody)

    @topspin said in In other news today...:

    While we’re going off-tangent, that reminds me of a situation in the supermarket last week. Unfortunately, because of word order this only works in German and not when translated, but I’ll translate it anyway.
    Supermarket radio: “Don’t want any of our coupons anymore...”

    My store does online coupons now. They send me emails a couple times a week or so and I login to my loyalty account there. They list all of the coupons and you can click on them to "load" onto your card.

    Of course I've made a bookmarklet to do that for me:

    javascript:$('.c-coupon__button').click();



  • @boomzilla One of my local drugstores used to do this. I think now all the offers that apply to you get loaded on your card automatically.


  • ♿ (Parody)

    @hungrier said in In other news today...:

    @boomzilla One of my local drugstores used to do this. I think now all the offers that apply to you get loaded on your card automatically.

    It's interesting, because there are things that are on sale every week if you have a card (or the cashier puts theirs in if you don't, quite often...man, what a scam that must be, since we get points towards discounts at local gas stations) that are marked that way in the store, but this stuff isn't. And the durations and amounts vary widely, so it would be a huge PITA to update the price tags.



  • @Cursorkeys said in In other news today...:

    First the FAA abdicating from its responsibility and now the FDA.

    That seems a bit misleading. That one-line summary makes it seem like they may be shutting down the reporting system so there's no way to report such incidents, but that's not at all what they're doing. The system they're shutting down is the alternative, non-public system that allowed these reports, and many others, to be hidden. All such reports will now have to be made through the primary, public system.


  • 🚽 Regular

    @HardwareGeek said in In other news today...:

    @Cursorkeys said in In other news today...:

    First the FAA abdicating from its responsibility and now the FDA.

    That seems a bit misleading. That one-line summary makes it seem like they may be shutting down the reporting system so there's no way to report such incidents, but that's not at all what they're doing. The system they're shutting down is the alternative, non-public system that allowed these reports, and many others, to be hidden. All such reports will now have to be made through the primary, public system.

    I personally meant abdicating their responsibility in the sense of having this hidden alternative reporting system in place at all. But that tag line does seem misleading until you read further.



  • @TimeBandit said in In other news today...:

    Very insurmountable. They'll have to use a whole 10-15 minutes to compile a more recent Linux kernel, with RISC-V support, before carrying on as usual. Instead of using the same old version from 2005.

    ...And then another 2-3 months to port all their drivers and crap onto the new architecture. Damage limited partly by Java, for once; positively riot-inducing.

    But since all the competition faces the exact same obstacles, they won't go under. It's just a 2-3 month profit-hit.


  • Banned

    Wolne sądy!

    "Read, Your Honor! - for hours and in empty court. Because that's the law."

    There's a verdict in Amber Gold case, but we don't know yet what it is. Only the judge knows for now, who will have to spend the next few months reading out over seven thousand pages of the verdict - because that's the law. The judge must read it - even if nobody will be there listening.

    Judge Lidia Jedynak started reading out the verdict on Monday. It can last even four months.

    Criminal Procedure Code gives very little leeway in shortening the reading. The judge is allowed to skip the charges, and the judge has chosen to do that. But it will still take a lot of time, because the judge altered one of the crimes that the Mr and Ms P. were charged with. In this case, she must read out all her findings and there's no way around it. The problem is, this charge is about crimes that affected 18 thousand victims.

    (Google Translate - contains great many funny mistranslations)


  • Discourse touched me in a no-no place

    @topspin said in In other news today...:

    The only real solution would be borrowing @dkf's cattle prod.

    Keep it. I can get more.


  • Discourse touched me in a no-no place

    @izzion said in In other news today...:

    It does take quite a while to remove all those ink bomb loss prevention tags from the clothing. And sign the customer up for the three new store brand credit cards and six mailing lists for new deals & coupons.

    Also, supermarkets optimise very much for throughput as they mainly trade in low margin necessities, whereas department stores count on larger profit margins per sale.


  • Discourse touched me in a no-no place

    @acrow said in In other news today...:

    And then another 2-3 months to port all their drivers and crap onto the new architecture.

    Redoing the masks for the new hardware will be the slow part, provided both the old and new architecture use the same memory mapping scheme.

    The actual bad part is that ARM is quite a lot more efficient overall (as in the total amount of energy to achieve a particular high-level operation) than RISC-V, and that's difficult to work around. The effect is particularly noticeable with ARM's commercial compiler, which has a bunch of tricks up its sleeve that aren't in either gcc or clang (indeed, clang is very unprepared to step into this area as it doesn't support the common C extensions for embedded work, and C++ isn't equivalent in this area at all).



  • @dkf said in In other news today...:

    @izzion said in In other news today...:

    It does take quite a while to remove all those ink bomb loss prevention tags from the clothing. And sign the customer up for the three new store brand credit cards and six mailing lists for new deals & coupons.

    Also, supermarkets optimise very much for throughput as they mainly trade in low margin necessities, whereas department stores count on larger profit margins per sale.

    I thought of that, but I am still continually amazed at how slow the lines in department stores often move. Yes, there would be greater motive to speed things up in supermarkets, but I would have thought that there would be some motive for the management to speed things up in a department store.



  • To everyone's surprise 🙄



  • @TimeBandit Yeah, and that's just the start of it. Just wait for when it gets to all those situations where the literal meaning of the road signs is either nonsense, or contradictory to their common sense meaning that humans have no trouble understanding. Temporary signs that override the permanent ones, but somebody forgot to blank or cross those is most common, but there are many strange combinations of permanent signs too. E.g. a town limits sign resets speed limit, but if there is a lower speed limit just before it, the signer almost certainly meant it applies beyond it. But if it is higher, it is just gradual speed reduction…

    (from the comment): When I moved from Texas to Oregon, I found out that the U-Turn laws are almost completely opposite

    When I cross the border to Austria, the speed limit behaviour suddenly changes. On our side speed limit is reset at junctions and town limits, but in Austria they have higher priority and only end when ended explicitly. But they are reset when leaving motorway, because it should be obvious to everybody that you can't go 110 outside motorway, right? And to be honest I am not sure which way they behave in Germany, because I didn't drive much outside motorways there and the sites about driving abroad fail to mention. It leads to some weird signage too—e.g. on a T-junction a sign across the junction with arrows to the sides indicating the speed limit on the road you are joining.

    Austria and Germany also have end of town limits and end of speed limit on the reverse of the corresponding start sign for the opposite direction, so they may be on the left. But that's a special case…



  • @Bulb Whenever someone explains the upcoming utopia of self-driving cars to me, I always think of the thousands of miles of rural highways in my old stomping grounds. No painted lines or reflective markers, no speed limit signs, no shoulders, oftentimes there are tractors and escaped cattle and herds of deer right in the middle of the asphalt, hills that will literally send you airborne if you go at the legal (but unposted) speed limit of 55 mph as well as sudden curves that you can't navigate at that speed...



  • @dkf said in In other news today...:

    Redoing the masks for the new hardware will be the slow part, provided both the old and new architecture use the same memory mapping scheme.

    A set of masks for a complex chip takes maybe a month. But before making the masks, they have to pull out the ARM core(s) and plug in the RISC-V (if that's what they're going to use).

    From what little I've found on RISC-V this morning (I've never used it, so I wasn't previously familiar with it), it doesn't use the AXI bus internally, but connects to the rest of the chip through an AXI bridge. Connecting the RISC-V to the rest of the chip is probably not prohibitively difficult. However, it's going to be time-consuming. It's a huge change in the chip architecture. I'd guess 6 months just to make the RTL changes. The bus timing of the bridge is probably not the same as what they have. The bandwidth available for various peripherals won't be identical. If they're really lucky, they might be able to do it in 2 or 3, but I doubt it.

    Once you have the implementation changed to use the new CPU core(s), you have to verify that the logic does what it's intended to do. Verification of a complex SOC may take 6 to 24 months. Some of that overlaps with the design/implementation time above, and the process is iterative as the verification process finds errors in the implementation.

    Somewhere, maybe about 6 months into this process, software people can start developing firmware/drivers based on a hardware-assisted software emulation of the chip. They can't finalize the software, though, until they get working chips in another 6–12 months.

    Repeat, although on a greatly shortened timeline, because the first version of the chip will have bugs, especially on a brand new architecture.

    Since the ARM ban applies only to new chips, a lot of that work will have to be done anyway. But changing the CPU core(s) mid-project is going to delay it by a few months, at minimum.


  • ♿ (Parody)



  • @dkf said in In other news today...:

    ARM is quite a lot more efficient overall (as in the total amount of energy to achieve a particular high-level operation) than RISC-V, and that's difficult to work around.

    Do you mean more efficient because of
    a. ISA differences
    b. longer time spent optimizing the ARM core for real-world workloads
    c. transistor level technological efficiency advantage
    d. all of the above
    e. something else entirely?

    particularly noticeable with ARM's commercial compiler

    Let me guess. Collecting discrete integer operations into SIMD batches á la Intel?



  • @acrow said in In other news today...:
    I don't know anything about any of your other points, but...

    c. transistor level technological efficiency advantage

    Very possibly. AFAIK, when you license a CPU core from ARM, you get a bunch of Verilog modules that are the implementation of the CPU. They are written and tested by ARM (and some parts may be encrypted; I've never dug that deeply into the CPU code). There are compile-time constants (think #define ENABLE_FOO 1) that let you choose, for example, between 32- and 64-bit data bus, the width of the address bus depending on how much memory you need, enabling optional features, etc., but ARM wrote and tested the options.

    You then build the rest of your chip around this core.

    RISC-V, OTOH, as far as I could find in my little bit of research this morning, defines the architecture and instruction set, but doesn't provide an actual implementation. There are several implementations, but they, or at least the one that seems to be the most popular in my limited search, is an automatic code generator, a level of abstraction above the implementation, not itself a concrete implementation. It, too, has options for address and data size and other features, including a choice between an in-order or out-of-order execution core. However, it is described as a base for academic and industrial research, which suggests to me that it may not be highly optimized. And since what it generates isn't a concrete, previously tested implementation, you're going to need to thoroughly test it yourself; that's going to take a bunch of time.

    As for transistor-level efficiency, for either core, that depends very much on the technology of the foundry you choose to manufacture your chip, and your physical design.


  • BINNED

    @mott555 said in In other news today...:

    hills that will literally send you airborne if you go at the legal (but unposted) speed limit of 55 mph as well as sudden curves that you can't navigate at that speed...

    Reminds of the posted speed limits of 80kph (as opposed to the unposted general limit of 100) when I visited Ireland. On roads best described as "that's not a speed limit, that's a challenge." (Not necessarily bad roads, but really narrow)


Log in to reply