It turns out having redundant datacenters for a cloud region doesn't help much if you apply the same (mis-)configuration change to all three datacenters at once.

Microsoft are reporting a misconfiguration in Azure that caused compute resources to lose access to Azure Storage (link may require an Azure subscription). This has caused a large number of Azure-based systems to fall over and MS are picking up the pieces now.

This section of the Brexit deal looks a lot like the technical Annex in Decision 2008/616/JHA from the Council of the European Union: https://ancillary-proxy.atarimworker.io?url=https%3A%2F%2Fwww.legislation.gov.uk...

There are a lot of respelling systems for English. I.T.A. strikes me as a compromise between traditional and phonetic spelling that's just plain confusing. For example, the diphthongs in "oil" and "owl" are represented with symbols that are both based on "o", even though there's no common vowel sound in "oil" and "owl" ("oil" and "aul", for example, would better illustrate the sounds, while sticking to existing English letters). The system also blithely conflates diphthongs and long vowels. The reverse "z" also obviously exists only to indicate a "z" sound spelled with "s" in English. All in all, I don't think it goes far enough toward being phonetic to benefit learners.

I think a reasonable approach would be to start with an existing phonetic alphabet and then remap the letters to match the English alphabet as close as possible. The Americanist phonetic notation, for example, uses "y" where the International Phonetic Alphabet uses "j". Going a bit further in assigning letters for convenience of writing English, you could use "j" for the "zh" sound in e.g. "vision" (making "judge" come out as "djâdj"). Vowels are going to need either new letters or a bunch of diacritics. I prefer the former, but Slashdot has fewer problems with the latter, and it would be very convenient if English could still be written in ISO Latin-1 (and supersets like Windows-1252).

In any case, I think the right approach is to start by using the reformed writing system as a learning tool and then gradually extend the areas in which it is used until the old system is fully supplanted. Experiences with I.T.A. suggest that introducing two different writing systems to young children is just confusing, so the switch to traditional spelling should be delayed until the child is comfortable with reading and writing.

Having to deal with the way non-native speakers mangle English on a daily basis, another option occurs to me: retain the spelling of English and change the pronunciation to match. This would make the morphological aspects that are retained in the spelling even more obvious. For people with more experience of written English, this would make the language easier to deal with.

We obviously wouldn't want to let good keyboards go to waste simply to save everyone a year or so learning to read. Would you be OK with memorising combinations for any additional letters (e.g. AltGr+Shift+D = "Ð") or do you need stickers to place on your key caps?

A large part of the reason why whole-language versus phonics is even an issue for English-speakers is that the correspondence between the spoken and written language is unnecessarily complex in English. Generally speaking, children learn to read and write much easier and better in languages that have a simpler relationship between writing and speech such as Finnish or Korean. If you can pronounce a word correctly in Finnish and know the alphabet, the way to write it is more or less obvious. Hence, Finnish doesn't even have a word for "spelling"; the closest equivalent is "oikeinkirjoitus", which simply means "writing correctly" (and arguably covers all aspects of orthography).

I'm tempted to say "Fiks ði inglish speling sist'm ænd kidz wil løn matsh isiø.", but the lack of a consistent standard pronunciation for English makes it difficult to agree on how to reform the spelling (you should be able to tell where I learnt to speak English). We may end up having to agree on how to speak English as a first step, or accept even more regional variation in spelling.

Also, there is a huge amount of written material in English that you'd want to retain access to. Automatic translation would render it possible to automatically convert Traditional English to Simplified English in many scenarios, though, so it might be possible to do a gradual transition where children learn Simplified English first; we could then gradually phase out the traditional spelling.

Furthermore, English has way too many phonemes for its alphabet, but many systems (like Slashcode) don't support non-English letters consistently enough (the "tsh" trigraph in my example was originally a small letter C with caron, but I can't convince Slashdot to show that) and you'd have to replace keyboards and other input methods if you want to add more letters. I suppose a compromise in which English spelling is simplified but retains the current alphabet would be workable.

I think you've misunderstood how Stadia works. The game is executed entirely on a device in a Google data centre. The only thing that is transmitted to the user is the output from the game (primarily video and audio). Barring leaks, only the developers and Google will have access to the executable and content files that make up the game. Even assuming the client device has a user-accessible cache, it would only contain the rough equivalent of a YouTube clip of your gameplay.

PC and mobile applications that have their own certificates and no option to add others are going to start breaking in Kazakhstan.

I also wonder how this is going to impact Internet-enabled devices that communicate with one particular service or company or for some other reason don't provide an option to add certificates. For example, how would this affect a video game console like a PlayStation 4? Do PlayStation Network services still work in Kazakhstan?

Actually, "Lýðveldið" is Icelandic for "the Republic". You could call Iceland "Lýðveldið Ísland" ("the Republic of Iceland"), but "Ísland" ("Iceland") would be the common name. Just calling it "the Republic" would be as silly as calling Great Britain and Northern Ireland "the United Kingdom" or America "the United States". Oh, wait...

Wouldn't it be simpler to have normal chairs and move the head and the tail of the queue, as in a circular buffer, rather than moving all the chairs? Of course, that would require the queue to be somewhere else than leading to the thing people are queueing for...

Considering the sheer amount of statistics being thrown around in what passes for Science, it might be more accurate to say that Science is the way we find things are likely to be true and likely to be untrue.

That said, I'm having trouble defining a representative sample of Science...

That's almost certainly a translation error. The University of Tampere press release states that "these studies clearly show that members of the group B coxsackieviruses are associated with the risk of type 1 diabetes", and the offending sentence in the Yle article would be the same in Finnish irrespective of whether the virus found is the only one or not (e.g. "löytänyt viruksen" would be "discovered a/the virus"). Finnish grammar doesn't have the concept of definiteness, meaning that a translator working from a Finnish source text would in many cases have to guess the intended meaning or look it up elsewhere. For similar reasons, many Finns have problems figuring out whether to use a definite or indefinite article when writing in English.

See Google Apps Status Dashboard for more details (hover over red outage dots for times).

I like your idea, but I find your vowel mapping very confusing. I think diphthongs should be spelt as their component vowels. Using the IPA for English summary on Wikipedia to enumerate the necessary combinations and noting that many vowels in English only come in short or long versions (and blatantly recycling the apostrophe), I'd suggest:

"a" as in "trap", "aa" as in "palm" or "start", "ai" as in "price", "au" as in "mouth"
"e" as in "dress" or "error", "ei" as in "face"
"i" as in "kit", "ii" as in "fleece"
"o" as in "lot", "oi" as in choice, "oo" as in "thought", "ou" as in "goat"
"u" as in "foot", "uu" as in "goose"
"'" as in "a" or "comma", "''" as in "strut", "nurse" or "hurry"

Consonants are mostly straightforward. "b", "d", "f", "h", "k", "l", "m", "n", "p", "r", "s", "t", "v", "w" and "z" all have obvious values. "g" is needed for "guy", so we use "j" as in "pleasure" (hence "dj" for "jam"). "y" is as in "yes". This allows us to reuse "c" for "th" as in "father", "q" for "th" as in "thigh", "x" for "sh" (and hence "tx" for "ch"). I'm not quite happy about having to use a digraph for "ng", but that should not be much of a problem for English.

Nau, yuu mei fiil cis luks sili, and aim x''r meni wud 'grii. Cat sed, geting juusd tu taiping laik cis is not ool cat haard. Sam mait aargjuu cat cis iz not inglix eni mo'r, b't ai qingk piipl xud 'dj''st priti kwikli.

Being able to type symbols like []/\;= with single key presses is a distinct advantage of the US layout over many European layouts when programming. I actually spent a few years programming with a US layout (switching back to Finnish when typing something in e.g. Finnish), but the difference is small enough that I felt that switching back and forth is not worth the trouble. Besides, the Finnish multilingual layout works for pretty much any European language with a Latin-based alphabet.

Many people don't seem to realise that the labels on PC keyboards' key tops are purely cosmetic; in most operating systems you can switch between layouts quite easily. As long as you don't need to look at the keyboard to know which key is which, you can easily use a different layout. In other words, switching a keyboard to a more familiar layout than the one it is labelled in works quite well.

If you work with several different languages with different alphabets, you are more or less forced to switch layouts as required by the current task. For example, a Greek programmer will almost certainly spend much of their time typing program code, commands or suchlike with a US layout (or similar) and switch to their local layout to type in their own language.