Just remember, this is not limited to official images, and it's not weighted in any way based on usage/downloads. So a crypto image that's pushed and never downloaded has equal weight with the ubuntu:18.04 image
I did not see in the article nor so far here in the comments one example of this in the wild, which perhaps indicates that using such simple sampling approach isn't common? If someone could successfully execute this against Twitter or Reddit, for example - that changes its newsworthiness completely.
Amazon have a pretty good alternative, so they'll only warehouse in Sweden if it makes economic sense vs Germany + Poland. As for the unions, their incentives and alternatives may be different to the union members too.
I always attempt to shop from Amazon Germany because (a) larger variety of items, and (b) I feel I can trust the reviews there. Have you found Swedish e-commerce sites to have enough reliable reviews to buy anything with confidence, or do you tend to buy products that you can find reviews of elsewhere? I joke to my wife, "If 500 Germans say it's good, I'm sold"
There are far fewer fake reviews at amazon.de (well, compared to what I read on HN) and surprisingly many that are high quality. I think it might be partially because the reviews would need to be in German so you have fewer low-cost ways to have them created.
And I’m not sure we have co-mingling here because I never heard of the problems that seem to be common in the US (/ on HN) with it and also never experienced it myself (I spend around 2.200 € on Amazon per year)
I live in the US and out of hundreds (thousands?) of purchases only one or two have made me question if they were counterfeit, and even those worked well enough and I didn't return them. There may be a selection bias in the comments you read because the countless people like me never post their experience and the 0.001% who unluckily were repeatedly sold counterfeit will likely post about it.
I think HN commenters also overstate the problem. I’m in the US, use Amazon a lot, and haven’t noticed counterfeits. Maybe I’m just bad at identifying them.
I think it’s the type of issue where it would be a major problem for the producer of a specific product that got ripped off, and any customers who are fans of that product, but may never affect the average buyer.
No idea. I haven't experienced that either and I use amazon.de a lot.
I've had one item sold by Amazon as new that was clearly a fraud return, though, i.e. the item inside the product package was switched for a different model.
That cluttering and dumping of tons of similar indistinguishable products under different sellers really annoys me. Another annoyance even when I do shop in those redundant product categories is that they don't even upload reallife photos of their products but these lowsy idealized CGI photos that give no sense to the quality or actual appearance of the item.
I feel like to stock co-mingling issue is far less pronounced in Europe compared to what the American crowd here on HN says. I live in Europe, and never got a fake product from Amazon.
I actually use Amazon DE/UK/US for to look at reviews but I more often than not end up purchasing from a Swedish e-tailer. There are some exceptions though, like when the price is just much lower on Amazon DE. I had to buy a label maker last week and the price on Amazon DE for a specific model was almost 40% cheaper than buying it from a local store in Sweden.
Same here (mostly buying books). I look at the reviews on Amazon.com, Amazon.co.uk and Goodreads.com (great because it has reviews of Swedish books as well), then order from Adlibris (mostly) and Bokus (occasionally). Often the prices have been about the same. Even if they weren't, I don't mind paying a bit more if it means the warehouse workers are treated fairly.
Yup, that's what I'm using as well, usually in combination with a check on ebay.de and ebay.co.uk for products which fit the bill - i.e. not for heavy, bulky goods like fridges and washing machines but certainly for smaller electronics and tools. For larger computer hardware - servers etc - I tend to look towards Germany and the UK since prices tend to be quite a bit lower than here in Sweden. I have only bought something at Amazon once, using the French version, my reasons being that they tended not to want to ship to Sweden and even if they do the price tended to be matched by other sellers. I do not want to get an Amazon monoculture so I'll continue to do so unless the price difference ends up so big that the offer is "too good to refuse". Swedish prices tend to be on the high side but they have come down, most likely due to competition from abroad. The Swedish postal service does its best to counteract this by putting a fine on parcels from outside the EU - they call it a handling charge but that is just word play, the intention is both for them to pad their budget and for you to stop buying from overseas - so I've been exploring ways to get around this by having goods delivered to the Netherlands (where I'm originally from) to import them myself when I happen to be there.
I don't think your point about the handling fee rings very true. If you followed the news at the time when it was added you would know that it was put in place to counter the huge workload that came with handling all the packages from cheap Chinese online stores (like Wish). Before that PostNord almost never charged for handling packages from outside the EU. The same can not be said about DHL, UPS and Fedex who ALWAYS make sure to charge extra for handling any package from outside the EU, almost without fail.
It makes sense in my mind that the buyer should be the one to pay for the handling of what they buy. I don't want to subsidise your shopping/shipping via my taxes, they're high enough as it is.
Postnord's argument that the handling of parcels from China took too much time was utterly nonsensical, especially given in the light of their usual complaints about the fact that fewer people were mailing stuff so they had to raise prices. If handling parcels from China took too much time they simply had not been very good in negotiating a deal with China Post about parcel rates.
Another factor in the nonsensicality of this tariff is that Sweden is more or less the only country which starts charging sales tax (called moms here, comparable with VAT in the UK, BTW in the Netherlands etc.) from a value of 0kr (i.e. free, but sales tax is paid on the shipping costs). Where other European countries start charging sales tax when the value of imported goods rises above (e.g.) €20, Sweden starts at €0. Many parcels ordered from China fall under that rate which make it a very common thing to be represented with the following:
price of ordered goods including shipping: €4
sales tax: €1
"handling fee": €8
total: €13, of which €9 goes to the Swedish state (directly or through Postnord)
Mind, this is not the shipping fee as that was already paid for in the €4 purchase price. If the same parcel were ordered to the Netherlands the customer would pay €4 and that's it, no import fees and no tariffs.
The so-called "handling fee" is nothing but an excuse to put a tariff on international trade for individuals, a way to make sure the fruits of globalisation stay out of reach of the public.
I use prisjakt regularly as well but mainly for the price comparison feature between retailers. The reviews on the website are decent as well though but there are typically a couple orders of magnitude fewer reviews on products than on Amazon DE/UK/US.
The question here is whether the reviews on prisjakt.nu are fewer because they're true reviews while those on Amazon often are fake. The prevalence of fake reviews - both positive as well as negative - has made it more or less impossible to rely on them as a guide.
As a German, don't trust the reviews, they're heavily manipulated. I guess at some point (e.g. top seller with thousands of reviews) manipulation doesn't move the needle as much, but for plenty of products with < 100 reviews, there's often dozens that are likely fakes (even without counting the paid Vine reviews).
Read 3/4 star reviews, they are usually good indicator of quality.
There are few type of those, and you are looking for someone who knows what they are talking about and talk about shortcomings of a given product. They are usually pro's that have some experience with given category of a product. So they know what they want.
If you find one like that its ease to judge what features you need vs features pros need.
Usually, 1 stars are by entitled complainers (unless product is cheap knockoff) and 5 stars are by people that don't know what they are talking about or brought reviews.
How do you feel you can trust Amazon reviews? In the UK they've become really bad, obvious fakes, loads of low quality products dropshipped from Alibaba/China with tons of 5 stars.
The only reviews you can vaguely trust are the non-5 star ones.
I don’t trust the reviews and the selection is such a grab bag of garbage. Besides some obviously branded stuff (I don’t know, any Apple product for example) it feels like playing Russian roulette, especially if you just want something random like, I don’t know, kitchen utensils.
I live in Stockholm, and order regularly from Amazon Germany already. It's free shipping if you order 39 euro or above, which usually delivers things in around a week. Fast shipping of two working days costs 9 euro per shipment.
Based on what I've read and this article, it doesn't sound like that much will change once it officially "launches" a Swedish site, although maybe Amazon will lower the minimum amount for free shipping or speed up regular shipping time. Re: this article, I can see them being willing to hold off for quite some time.
Also to note: During the Covid-19 peak in Sweden, Amazon provided to-the-door delivery using Postnord, which I figured was somewhat of an experiment for when the Swedish site launched. It's pretty rare here in e-commerce to get door deliveries - most online shopping is sent to local convenience stores and you need to go and pick up.
Something I forgot to add: Sweden does have some great gig worker to-the-door delivery companies that have sprung up, particularly around small items like from online pharmacies. When you fulfill an order using those companies they typically have a great monitoring app to track where your delivery is, to input your building door code if necessary, inform you when it's been left outside your door, etc.
I always find it a great contrast to compare to premium shipping companies like DHL and FedEx, whenever I'm unlucky enough to get a delivery from them from overseas. They're still stuck in a world of "we'll deliver between 9am and 6pm, be ready" and they also make little attempt to ask for door codes if you live in the inner city, so most first attempts end up with "premises was not accessible" with no convenient way to give them the four digit code they need. I spent 1+ hour on hold last week with DHL and FedEx for this exact purpose. It will be interesting to see how "modern" of an approach Amazon takes.
Germany has interesting approach for this: if you can't be reached, the delivery will be left to whichever neighbour or coffee shop happens to accept it. You'll get notified where it was left, and then just go knocking.
It's quite weird at first but in my experience works surprisingly well. Very rarely one has to go queue in the post office to collect the parcel.
That’s only the case if it was delivered by Deutsche Post (instead of DHL) and does not fit into your mailbox, or, in the case of DHL, when you told them to do it via their website.
I agree! Budbee in particular has been very good in this respect! Often you can get free shipping with delivery between 17:00 and 22:00, and if you want a specific hour, you can pay 39 SEK to get that.
I upgraded to version 3 but feel like they haven't made a product improvement since, despite the new subscription model. Staging of chunks still remains the killer feature to keep me with it but I'd be happy to switch (pay) if anything else came along.
Benedict Evans doesn't seem like the kind of guy to fall for BS though, let alone defend it. And surely at least one of their investors had the technical due diligence capability to investigate how much computing power is in the demo "rigs" and be able to know if such power can be "miniaturized" into a consumer product by year X, perhaps allowing for a little software optimization on the way too.
As it sounds like it's running inside Docker itself, I'd like to know if this supports "Docker in Docker". My requirement is not strictly DinD however I run multiple containers during CI (Postgres, node, test containers, etc). Possible via different approaches in CircleCI, Shippable and SemaphoreCI. I don't actually build any containers to save.
This is actually how Codeship's Docker infrastructure works by default, builds up containers and executes all commands in them natively. DinD still possibly but somewhat yet useful with that approach.
