Archive for the ‘USA’ category

Online advertising dollars: Google’s cut from your work is 40 per cent


From Bob Hoffman’s The Ad Contrarian newsletter of May 24: ‘two weeks ago a study by the ISBA and PcW that reported that half of every “programmatic” ad dollar is scraped by adtech middlemen’ and ‘According to a paper written by Fiona Scott Morton, an economist at Yale University, Google pockets about 40¢ of every online ad dollar before it ever gets to a publisher. Not just search dollars, not just programmatic dollars, but all online ad dollars.’ Just one more reason I refuse to sign these:

   I’m not part of the 90 per cent. And the bastards at Google are rich enough. Let them share it with illegal content mills as they are peas in a pod. Another solution for legitimate publishers is dearly needed.
   At least there’s been some sort of work with the commissions agencies take in other media, and that’s typically at 15 per cent here. Google is taking the piss with its automated systems.
   We know the US doesn’t have the balls (or funding?) to take them on at this point, but how about other sovereign territories in which Google operates? Surely they have to comply with our laws, too?

Tags: , , , , , ,
Posted in business, globalization, internet, publishing, technology, USA | No Comments »

Live from Level 3


Finally, a podcast (or is it a blogcast, since it’s on my blog?) where I’m not “reacting” to something that Olivia St Redfern has put on her Leisure Lounge series. Here are some musings about where we’re at, now we are at Level 3.

   Some of my friends, especially my Natcoll students from 1999–2000, will tell you that I love doing impressions. They say Rory Bremner’s are shit hot and that mine are halfway there. It’s a regret that I haven’t been able to spring any of these on you. Don’t worry, I haven’t done any here. But one of these days …

Perhaps the funniest Tweet about the safe delivery of the British PM and his fiancée’s son, for those of us who are Clint Eastwood fans:

Tags: , , , , , , , , , , , , , , , , , , , , , , , , , ,
Posted in China, culture, France, globalization, Hong Kong, humour, New Zealand, politics, publishing, Sweden, UK, USA, Wellington | No Comments »

Finding an Android browser that works without fuss is harder than you think


With my last two cellphones, I’ve not used the default browser. I usually opted for Firefox, and in December 2018, I believe that’s what I did on my then-new Meizu M6 Note.
   I don’t recall it being too problematic, but the type on some sites displayed a tad small, so I sampled a few others. I must have tried the usual suspects such as Dolphin and definitely recall seeing the Brave icon on my home screen, but my friend Robin Capper suggested Edge.
   You might think that that’s a ridiculous option given what Edge’s (and IE’s) reputation has been like, but it actually worked better than the other browsers I sampled. It played the videos I loaded on it, and it displayed type generally well, but there was one very regular bug. If I left a session and came back to it later, or let the phone go to sleep or standby, Edge would almost always falter when I tried to pick up where I left off. It would stutter and close. When I opened it up again, it was fine.
   The latest version began displaying in my notifications that it wouldn’t work properly without Google Services, which was a blatant lie, since it was still stable other than the bug above, and all previous versions were absolutely fine. I wonder if this was some leftover from the Chromium base, but, as with the overwhelming majority of Android apps, Google Services are unnecessary.
   The other bug that began happening on a more recent version was Edge getting confused by stylesheets and not knowing what size to display type at. It might change as you browsed, and when you scrolled back up the page, the text that was legible before had turned minute. It did this on Lucire, and it is serious enough for us to redevelop a template for the site.
   I began wondering if there was life outside Edge. I returned to Firefox to find it stable but utterly incapable of playing videos. I don’t remember it being like this when it was my default, but like so many software programs, the more they upgrade, the crappier it gets. I also believe that a lot of these boffins don’t test with older gear, for reasons I’ve outlined elsewhere on this blog.
   Four browsers were suggested to me as replacements: Vivaldi (which I went to anyway, since I use it on the desktop), Duck Duck Go (which I had heard was slow, but I downloaded it anyway), Brave (they have a programme where they claim to give money to publishers but it’s impossible for a publisher like me to sign up to), and Bromite (hadn’t heard of it before today). I had already tried, and rejected, UC Browser on another occasion.
   Vivaldi has been and gone from my phone as I write this post. It’s buggy as heck. Twitter displays about half a centimetre off, so you think you’re clicking on one thing you see on the screen but you’ve just activated the link that’s 0·5 cm above. YouTube will crash the browser (two out of four times). It loses the tab you were browsing on when you come back to a session. It gives the same BS about needing Google Services when it doesn’t. I was very disappointed considering it syncs with Vivaldi on the desktop, the settings seem comprehensive, and the interface looked pretty good.

Vivaldi struggles to display YouTube before crashing

Vivaldi displays everything a bit low (though it functions as though everything is fine, leading you to click on the wrong things), and the tabs I set it to show have gone

   Duck Duck Go has been working quite well. Other than the pop ups that tell me about things I already know as a decade-long user of the search engine, I haven’t noticed the slowness that I’ve heard from a very reliable and knowledgeable source.
   Brave was back, still telling me about their rewards’ programme, but I haven’t experimented with it enough to form a proper opinion. But it has sent a notification about my first Brave advertisement, which I actually can’t see. I admire what they’re trying to do but if only they’d let me sign up as a publisher—yet their site doesn’t permit it. It might be short-lived on my phone, too.
   Bromite, so far, has worked in a standard fashion with nothing too remarkable, and I’ll be investigating further.
   The day has ended rather differently on the cellphone—a whole lot of time invested on a device I barely use. But it’s been a fun exploration of what’s out there and how some fall well short of the basics of stability, consistency and compatibility. Duck Duck Go has so far won the default slot but the jury is still out on Bromite.

Tags: , , , , , , , , , , , , ,
Posted in cars, China, design, technology, USA | No Comments »

Facebook still can’t get the basics right


For a guy who gave up updating his own Facebook in 2017, and uses it just for work stuff, I’m still amazed at how many bugs I come across.
   Two days ago: discovered that you can’t post links. The ‘Publish’ button is greyed out.

   Yesterday: I wanted to tag one of our writers on Lucire’s Facebook page. Don’t think it worked but the other thing that didn’t work was the link preview. This is an old bug and I remember it from when I used Facebook regularly, so things must move really slowly there. You can post links again—sort of. Basically, the posted link and the link in the preview have different URLs. The one in the preview takes you to a 404: it’s the correct URL with the author’s Christian name appended to it, to make it wrong. I’m glad that for the most part, I leave this page to automation—it’s actually more accurate than going in there and posting directly into Facebook! You would think the opposite would be true.

   What is good is that you can delete posts now, which you couldn’t three weeks ago.
   Today: they’ve got rid of the news feed, which is actually a good thing, but I know how others like it. I went in just before I wrote this post in case it was a fleeting bug this morning, but there’s still no news feed.

   I’ll might look again next week if work stuff comes up. Three visits like that, one or two a day, is anomalous for me these days. All these visits showed is that Facebook is no less buggier than it was half a decade ago, with pretty much the same bugs: regular failures in posting, linking, and displaying databased content. In fact, it may be worse as the whole thing appears to crumble under its own weight.

Tags: , , , , ,
Posted in technology, USA | 2 Comments »

COVID-19 stats’ update, April 16


Don’t worry, I won’t make this too regular, but as I had done some more number-crunching of the available stats during the daytime, I thought I’d share them. I’ve noticed that some countries update their test numbers on a less regular basis, e.g. France, Singapore, Sweden and Switzerland, though Worldometers now has updated ones since my last COVID-19 post. France’s test figure hasn’t changed, so we can safely conclude that its infection rate as a percentage of tests done is lower than what’s cited below. The same applies to Singapore.
   New Zealand has dipped below 2 per cent, finally, but thanks to rounding it’s cited as 2·00 per cent below. These figures include what Dr Ashley Bloomfield announced an hour ago. Happily, the US has started to see a fall since I last did these figures—there’s one post I didn’t write even though I had the calculations ready (it was too late at night for me to compose something cogent). Goes to show how quickly the landscape changes.
   I had overestimated the number of tests Sweden had done: it turns out they haven’t increased in number at the same rate as the fortnight before, though my use of 75,000 in the previous table wasn’t far off. Despite my overestimation, their infection rate continues to rise.
   The UK has also risen but not at the same rate, though judging by Twitter there, some are questioning whether deaths in aged care facilities are being included.
   Germany should be happy with its rate going from the 9s into the 7s.

France 147,863 of 333,807 = 44·30%*
Spain 180,659 of 650,755 = 27·76%
UK 98,476 of 398,916 = 24·69%
USA 644,089 of 3,258,879 = 19·76%
Sweden 11,927 of 74,600 = 15·99%
Italy 165,155 of 1,117,404 = 14·78%
Switzerland 26,336 of 199,000 = 13·23%
Germany 134,753 of 1,728,357 = 7·80%
Singapore 3,699 of 72,680 = 5·09%*
KSA 5,862 of 150,000 = 3·91%
New Zealand 1,401 of 70,160 = 2·00%
South Korea 10,613 of 538,775 = 1·97%
Australia 6,462 of 377,024 = 1·71%
Hong Kong 1,017 of 116,273 = 0·87%
Taiwan 395 of 49,748 = 0·79%

* Test number has not been updated

Tags: , , , , , , , , , , ,
Posted in France, New Zealand, Sweden, UK, USA | No Comments »

COVID-19 infections as a percentage of tests done: April 13 update


I can cite these COVID-19 calculations (infections as a proportion of tests done) with a bit more confidence than the last lot, where many countries’ testing figures had not updated. I see the Kingdom of Saudi Arabia has released its total test numbers now, and they show a pretty good result, too.
   Compared to my post of the 7th inst., there are improvements in France, Italy, Switzerland and Germany, while Spain has shown a marked and positive improvement (from 39·58 per cent to 28·25 per cent).
   The UK’s delay and its initial reliance on herd immunity, with sycophants up and down the country agreeing, is showing up now as its number grows slightly, from 20·4 per cent on the 7th to 23·88 per cent with the latest data.
   The US’s numbers are holding fairly steadily with an increase of 0·8 per cent since the 7th (to 19·78 per cent).
   Sweden’s total test figure is one of two inaccurate ones here, having remained unchanged since the last tables, which obviously cannot be right. I estimate they have done around 75,000 tests so far, which would bring the figure to 13·98 per cent, fairly close to the 7th’s, rather than the 19·16 per cent that the Worldometers’ table would have me calculate.
   Also statistically similar are Switzerland, South Korea, Australia and Hong Kong, though Hong Kong’s total test figure is also inaccurate (unchanged from the 7th). Singapore is showing a rise with the reports of community transmission. New Zealand is showing a small drop (2·71 to 2·15 per cent), though the percentage change here is less than what the US’s is.
   Taiwan continues to see its percentage decline with another 8,000 tests done and only an additional 17 infections since the 7th’s post.

France 132,591 of 333,807 = 39·72%
Spain 169,496 of 600,000 = 28·25%
UK 84,279 of 352,974 = 23·88%
USA 560,433 of 2,833,112 = 19·78%
Italy 156,363 of 1,010,193 = 15·48%
Sweden 10,483 of c. 75,000 = c. 13·98%*
Switzerland 25,449 of 193,800 = 13·13%
Germany 127,854 of 1,317,887 = 9·70%
KSA 4,462 of 115,585 = 3·86%
Singapore 2,532 of 72,680 = 3·48%
New Zealand 1,349 of 62,827 = 2·15%
South Korea 10,537 of 514,621 = 2·05%
Australia 6,359 of 362,136 = 1·76%
Hong Kong 1,010 of 96,709 = 1·04%*
Taiwan 393 of 47,215 = 0·83%

Tags: , , , , , , , , , , , , , , , , , , , ,
Posted in China, Hong Kong, New Zealand, Sweden, UK, USA | No Comments »

Is Facebook lying to customers about who has seen their ads?


Not withstanding that I can’t edit my advertising preferences on Facebook—they took that ability away from me and a small group of users some time ago (and, like Twitter, they are dead wrong about what those preferences are)—I see they now lie about what ads I’ve seen and clicked on.
   I can categorically say I have not seen an ad, much less clicked on an ad, for the US Embassy.
   It’s pretty hard for a person who doesn’t use Facebook except for work to have clicked on any ads on their platform.
   And as I’ve largely quit Instagram it’s highly unlikely I accidentally swiped and clicked on an ad there, too.
   On the remote chance that I did, then it shows that either Facebook’s or the US Embassy’s targeting is appallingly bad since I’m not American. I doubt that the US Embassy would have had such a wide market as to include me.
   I theorize, and I do so with zero proof, that Facebook is so deep in its con to claim certain advertising reach numbers that it’s falsely attributing hits to random users across the site. These may have been hits done by bots—bots that it endorses, incidentally—and now they want to pin them on legitimate people.
   It’s a hypothesis but given that I’ve been right about a few way-out ones (false user numbers, bot epidemics, malware downloads), I’m going to advance it. Now let’s wait four years for this to blow up into something.

Above: The only way I can view my advertising preferences on Facebook is through the mobile version. But here they cannot be edited. (The web version won’t show them at all.) They are also quite wrong that these are my interests, but since when have they been right anyway?

Tags: , , , , , , , , ,
Posted in business, internet, technology, USA | No Comments »

Are you doing on Facebook what Facebook does on Facebook? They’ll sue you


Pxfuel/Creative Commons CC0 1·0

Here’s quite a funny one for you this Easter weekend: Facebook apparently has filed suit against companies that do the following, according to Social Media Today.

• Companies that sell fake followers and likes, which Facebook has pushed harder to enforce since New York’s Attorney General ruled that selling fake social media followers and likes is illegal last February
• Two different app developers over ‘click injection fraud’, which simulates clicks in order to extract ad revenue
• Two companies over the creation of malware, and tricking Facebook users into installing it in order to steal personal information

In other words, Facebook has filed suit against people who do things that are variations of what Facebook itself does.
   The first. This has long been proven by Veritasium, and one would hope the defendant points out that Facebook has endorsed such behaviour, and that its terms and conditions have generally meant squat. Facebook allows hate groups (hate speech is ‘counter-speech’, they tell me), hates drag queens and kings, drags its heels in removing illegal content (eight clips of the Christchurch massacre are still on there, a year later), and preserves bots, fake accounts and phishing pages, all contrary to what their own terms and conditions say. These happen with such frequency that one might say they are Facebook policy.
   Now, Facebook mightn’t do the second but it certainly extracts ad revenue from customers, and not necessarily fairly. Click fraud? How about audience fraud? That’s been the subject of lawsuits against it. We’ve gone through this before on this blog, least of which is Facebook’s lying about its user numbers. It cites heaps of people but we know among them are bots; and we know that it claims more people in certain demographics than there are people. I’ve said this for a long, long time.
   Third: Facebook tricked users for years into installing a ‘malware scanner’ with purposes it would not go into. But it essentially admitted their scanners collected data from users (as reported in Wired, ‘Facebook tells users when they agree to conduct the scan that the data collected in the process will be used “to improve security on and off Facebook”’—it seems reasonable to conclude this is personal information). The scanner never appeared in one’s installed programs’ list, either, and in my case, knocked out my real antivirus software. We also know that when Facebook accused certain people of having malware, the company was lying. The scanner took a long time to run, so what was it sending back to the mothership? Conclude from all of that what you will, but tricking Facebook users into installing software that is hidden on a user’s PC and takes data off it is right out of a fraudster’s playbook.
   Given the amount of crooked activity that Facebook itself engages in, and the lies its team tells, criminals would be forgiven into thinking that it was a website that collected and ran scams, and that Mark Zuckerberg was a kindred spirit.
   The hypocrisy remains strong at Facebook.

Tags: , , , , , , , , , , ,
Posted in internet, New Zealand, technology, USA | 1 Comment »

Another COVID-19 table: total infections as a proportion of tests done


Peter Lambrechtsen rightly pointed out that COVID-19 per capita infection statistics aren’t as good as knowing the infection rate based on tests done, so at 2 a.m. I decided to crunch some numbers based on the stats I had on hand. These are many hours old now but hopefully still indicative of where things stand. Here you want a low percentage, and we are very fortunate to be sitting on 2·71 per cent. This site has tests per million as well, which I haven’t factored in. Taiwan and Hong Kong are looking even better on this measure; Australia isn’t looking too bad, either. The European and US numbers are sobering. Mainland China and the KSA haven’t released their testing numbers, only total infections.
   I don’t really want to go into fatality rates.

France 98,010 of 224,254 = 43·70%
Spain 140,510 of 355,000 = 39·58%
UK 51,608 of 252,958 = 20·40%
USA 369,179 of 1,941,052 = 19·02%
Italy 132,547 of 721,732 = 18·37%
Sweden 7,693 of 54,700 = 14·06%
Switzerland 22,242 of 167,429 = 13·28%
Germany 104,199 of 918,460 = 11·34%
New Zealand 1,160 of 42,826 = 2·71%
South Korea 10,331 of 461,233 = 2·24%
Singapore 1,375 of 65,000 = 2·12%
Australia 5,908 of 310,700 = 1·90%
Hong Kong 936 of 96,709 = 0·97%
Taiwan 376 of 39,011 = 0·96%

Tags: , , , , , , , , , ,
Posted in China, France, Hong Kong, internet, New Zealand, Sweden, UK, USA | No Comments »

One more COVID-19 post: graphing and animating the data


Russell Brown linked this COVID-19 trend page by Aatish Bhatia on his Twitter recently, and it’s another way to visualize the data. There are two axes: new confirmed cases (over the past week) on the y and total confirmed cases on the x. It’s very useful to see how countries are performing over time as it’s animated, and to get a handle on what trajectory you’re on.
   I’ve plotted us against some Asian countries and territories in the first graph and western countries in the second. South Korea is doing quite well and Taiwan is really bending its curve down. Try it yourself by clicking on either of the screenshot graphs below.

Tags: , , , , , , , , , , , , , , , , , , , ,
Posted in design, globalization, Hong Kong, New Zealand, Sweden, technology, UK, USA | No Comments »