Re: AI transcription and translation improves massively
Is it good enough to output text about a WWII fighter aircraft called the Cutler 109?
1452 publicly visible posts • joined 10 Jun 2009
> it is far deeper
I agree. This is in the uncanny valley between biochemistry (e.g. looking at enzyme-catalyzed organic reactions) and biology - examining the structure and behaviour of living organisms. I suppose that what I see in that video is just the complex cell wall structures organizing themselves into local energy-minimum conformations that resemble the assemblers. It's just very advanced crystallization. I guess it might have looked that way in the Primordial Soup, and look where that got us.
If I recall correctly, progression up the rating scale went in powers of two of the kill count, so it took just as long to go from Dangerous to Deadly as it did to go from Harmless to Dangerous. Well, I suppose one's kill rate increased, but not enough to make it possible for me to get past Deadly on a C64.
I've mentioned before that OOLITE does quite a good job of recapturing the feel of Elite while expanding its horizons a bit.
> what can they hope for in the ruling?
The legal process isn't like a bran tub where you pay for a ticket and dip for a prize. The original complaint must state the relief sought from the court. As nearly as I can tell "WhatsApp sought an injunction restraining NSO from accessing WhatsApp’s servers, violating WhatsApp’s terms, and impairing WhatsApp’s service. WhatsApp also sought compensatory, statutory, and punitive damages."
Source: Case: 20-16408, 12/16/2020, ID: 11930616, DktEntry: 32 [eff.org PDF]
Of course, it's in the interests of exactly zero office software suppliers to make this happen. I suppose the ISO could consult and specify a data interchange format, but you can be sure that the implementations of the import/export processes would variably mangle the data.
I am reminded of a project in the MoD [1] to integrate a number of commercial off-the-shelf programs into a comprehensive project management suite, two of which (for reasons that will not become clear again any time soon) were Microsoft Project and (as I recall) a PM program from Artemis. We had to have a rule that managers would not export/import between the two, because the two programs fundamentally disagreed about how a calendar worked and what a task duration data point meant; a data round-trip between programs resulted in a project schedule that was irretrievably broken.
[1] Many years ago, now.
> find yourself having to recompile the kernel
This is a myth. I've tried out many distros over the years, and I can't remember the last time I had to recompile a kernel. I've had to compile a module for the WiFi dongle that I stupidly bought without checking to see if it had extant support, but that was my fault.
And even if you did have to recompile the kernel, you are extremely unlikely to have to edit a makefile. make menuconfig doesn't require dev skills, it just needs you to know what hardware you're rolling for.
I've been using Waterfox Classic on KDE/GNU/Linux for several years, exactly because it supports the extensions I have grown to expect as part of my browsing experience.
As far as updates go, I find that when a new version is released, the update checker flags up for me that Waterfox cannot download the latest version. That's not a problem, really, it just means that I have to visit the website and grab the compressed tar file - as someone pointed out, if it's installed system-wide (mine lives in /opt) then root has to unpack the file, which is the sum total of "installation".
The other irritant is that Waterfox will (sometimes!) state on startup that it's not my default browser: it really is, and I've tried all the KDE settings I can find to confirm that, but WF isn't convinced. However, I can live with it, and I'll stick with it while I can. I can't remember the last time I lit up Firefox.
This was a fate somewhat narrowly avoided by Japan, too. The US Education Mission to Japan (1946) recommended that kanji, the characters adapted from Chinese writing, be replaced by romaji, the orthography of Japanese written with a Latin alphabet. This recommendation was based on little and outdated knowledge by the Mission members, ignored their terms of reference, and would have "invited the Japanese people to commit cultural suicide".
Ref.: The First United States Education Mission to Japan [pdf]
> why do we hear nothing about it in the public debate ...?
Some useful technical discussion and onward links in this source:
Technically, the Diffie-Hellman key exchange can be used to establish public and private keys. However, in practice, RSA tends to be used instead. This is because the RSA algorithm is also capable of signing public-key certificates, while the Diffie-Hellman key exchange is not.The ElGamal algorithm, which was used heavily in PGP, is based on the Diffie-Hellman key exchange, so any protocol that uses it is effectively implementing a kind of Diffie-Hellman.
As one of the most common methods for safely distributing keys, the Diffie-Hellman key exchange is frequently implemented in security protocols such as TLS, IPsec, SSH, PGP, and many others. This makes it an integral part of our secure communications.
Source: https://www.comparitech.com/blog/information-security/diffie-hellman-key-exchange/
I think everything that one needs to know will be in Domestic CCTV systems - guidance for people using CCTV [ico.org.uk]
That does not seem to place other people's property off limits, but says "[...] if your system captures images of people outside the boundary of your private domestic property – for example, in neighbours’ homes or gardens, shared spaces, or on a public footpath or a street [...] [t]hen the General Data Protection Regulation (GDPR) and the Data Protection Act 2018 (DPA18) will apply to you, and you will need to ensure your use of CCTV complies with these laws.
The analogy between <fashion item> and <book> is that there exist saleable copies of copyrightable prototypes of both items. The wedding gown could very well be a copyrightable design, and people who want to have copies need to license the design. Similarly, publishers have to do a copyright deal with authors in order to sell copies of the book.
I suppose that the OP was remarking on the curious situation that a novel clothing design does not become 'a fashion' until it has been copied extensively so that many people can wear it. The book analogy would not be the situation where a book becomes a runaway best-seller, more as if a thousand authors started writing stories using the original characters and scenario. Fan fiction, come to think of it.
> if you're deep enough there is no light to see things, so there is no looking around
Another thing is that the thick steel pressure hull, and then the surrounding casing, are both pretty much 100% opaque to visible light, even if there's a lot of it outside. Naval subs don't have portholes.
One's super-carrier does have guns, but not the sort of naval weapon designed to take on other vessels, or shore bombardment. There are some miniguns, and a close-in weapon system for last ditch defence, but if something has got close enough to the carrier to be shooting at it, then the destroyer screen has failed. Perish the thought. HMSQE potency is almost entirely within the aircraft.
@Neil Barnes
> ...actual person?
That's the exact meaning of "natural person", a term used to differentiate them from "artificial persons". The Legal Information Institute says
Artificial PersonAn entity established by law and given at least some legal rights and duties of a human being. Corporations are the most common types of artificial persons.
By that definition, AI systems (not being established by law) are not any sort of person, which seems to me to accord with common sense.
All orbits are horizontal, in the sense that they're heading towards (and over!) the horizon. The only way a polar orbit is "vertical" is from the point of view of someone looking at an Earth map with a north-up convention (although south-up would also work, but you don't see many of them).
Just to clarify: the doctor in question submitted forged Covid PCR test results at the airport, not vaccination certificates. Same offence, different documents. I am not an immigration lawyer, but should she apply to remain in the UK at some point, I believe the criminal record will count against being successful. In fact, a criminal conviction (beyond reasonable doubt) is not even necessary: the HO only sets itself a balance of probability standard:
A person will not normally be considered to be of good character if there is information to suggest that any of the following apply:Criminality
If they have not respected or are not prepared to abide by the law - for example, they have been convicted of a crime or there are reasonable grounds to suspect, meaning it is more likely than not, they have been involved in crime.
Source: Nationality: good character requirement [pdf]
Maybe someone thought that robots.txt would be the answer to keeping the pages away from Google? I have to say that I thought so, right up until I read:
A robots.txt file tells search engine crawlers which URLs the crawler can access on your site. This is used mainly to avoid overloading your site with requests; it is not a mechanism for keeping a web page out of Google. To keep a web page out of Google, block indexing with noindex or password-protect the page.
Source: https://developers.google.com/search/docs/advanced/robots/intro
Yep, I see the Joke Alert. The Print page isn't plain HTML, though:
jonathan@Odin:~$ curl -s https://www.theregister.com/Print/2021/07/20/pdf_html_debate/ | grep -A 1 "<script"
<script>
var RegArticle={id:216169,pf:0,af:0,bms:0,cat:'news',ec:['adobe'],kw:[["software",'Software'],["web",'Web'],["development",'Development']],short_url:'https://reg.cx/40A4',cp:0,noads:[],author:'Thomas Claburn'}
--
<script>var RegTruePageType = 'www print';</script>
<link rel="canonical" href="https://www.theregister.com/2021/07/20/pdf_html_debate/"><link rel=stylesheet href="https://fonts.googleapis.com/css?family=Arimo:400,700&display=swap">
--
<script>
var RegCR = true;
--
<script src="/design_picker/4c219a18bc536a8aa7db9b0c3186de409fcd74a7/javascript/_.js"></script>
<script async onerror="gpt_js_errored()" src="//securepubads.g.doubleclick.net/tag/js/gpt.js"></script>
<script>
RegGPT('reg_software/front');
--
<script async src="https://www.googletagmanager.com/gtag/js"></script>
--
<script>
(function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){
What are Apple and SIS if not aggregated employees? Nothing is done except by employees of these firms; firms cannot lie, only their employees. IANAL but I firmly believe this is an ancient principle of the law: masters are responsible for the actions of the servants in their employ. Minimal research leads me to the insight that there is a distinction between employer::employee and principal::agent relationships, but today I am avoiding rabbit-holes.