Posts by that one in the corner

In the battle between Microsoft and Google, LLM is the weapon too deadly to use

Re: Futile

> I foresee people feeding the AI with all the known math and physics knowledge and asking it to contemplate a theory that unites quantum and General Relativity,

I hope when you sat "the AI" you aren't referring to the LLMs that the article is talking about, as they aren't any good at simple logic and arithmetic problems, except by luck (as in, they had already seen that question and answer pairing in their training data and could repeat it back). Many places discuss the existing models failing, for example:

https://ai.stackexchange.com/questions/38220/why-is-chatgpt-bad-at-math

New models of IBM Model F keyboard Mark II incoming

Come down the rabbit hole, there is nothing to fear. We have cookies[1]!

[1] best avoid the little cakes.

Strange how unlike the Model F the layouts are

IMO, The XT version of the Model F has the best key layout, with a decent cursor/home/end cluster overlaid on the numeric keypad and the function keys easily accessible with a slight sideways motion.

I've got a collection of Northgate Omnikey Ultras (which are showing signs of aging: have to gird my loins and unsolder the S key on the current one), simply because they are the closest to the proper Model F layout I could find: none of that ghastly new-fangled inverted-T for the cursor keys, plus the ability to put CTRL back above shift where it belongs (you don't need a CapsLock after you've gone past two-finger typing on a computer keyboard, it was only there because holding down the shift on a classic workhorse typewriter was tiring!). Under Win'10 I've been trying out having the CapsLock mapped as the Windows key (the Ultras predating that), although that key is so rarely used I keep forgetting it is available now!

An XT Model F layout would be great to have. I'd even forego F11 and F12. Just not sure about the price (if I was still working full-time it would be a no-brainer to get a couple, but as things stand they'd have to be less expensive). Although I would miss the added lethality of the Northgate keyboards.

Google denies Bard trained using OpenAI ChatGPT responses

The inevitable xkcd

What goes around, comes around, just insert Bard, ChatGPT and the rest into this loop and Bob's You Uncle[1]

https://www.explainxkcd.com/wiki/index.php/978:_Citogenesis

[1] last week he was your great grandfather, but after updating his entry based upon this new information posted over here...

Re: "wrong to train its AI search chatbot Bard on text generated by OpenAI's ChatGPT"

So if I wanted to claim I had the Ultimate AI that had re-processed all the text available, all I *actually* have to do is change my Eliza to print out

"Send three and fourpence, we're going to a dance".

But to make sure it is AI, I'll use the proper language: I've a copy of Winston and Horn, 3rd edition, in the post.

OpenAI prohibits people from using its outputs to train their own models.

But they can just suck up whatever *they* want from anything published on the web?

Hypocrisy much?

After 11 years, Atlassian customers finally get custom domains ... they don't want

Re: Have some sympathy

In Australia, the bugs have you.

School principal resigns after writing $100,000 check to Elon Musk impersonator

Re: "I am very smart"

> For romance scams, you could well be contacting them on a dating site.

The trick is to make the mark *think* they are being contacted via the dating website, without actually doing so (as that would leave a trail - and mean the scammer has to pay the dating site fee as well).

So, again, just blast out many, many messages that claim to be a response to the mark's post on the dating site - almost all of the recipients won't have gone anywhere near the dating website, but a few will have - and a few more may fall for the "your friends signed you up" second email.

Bit of standardised social engineering later and kerching.

Business deals and investment[1] purchases of sums like that occur every day without doing all the paperwork via lawyers; if you feel like it, you can go and buy a car for that much without invoking the lawyers. If you have signatory power up to $50k already then signing big cheques can be normalised.

Agreed, if the deal is something out of the ordinary (for you) then getting a lawyer to go over the contract is a sensible precaution, but not a strict necessity (either way, as signatory, you still bear the liability - getting lawyers in just gives you someone else to blame).

But, of course, the whole way any of these scams work is to hope to find someone who doesn't go to lawyers as a matter of course, especially after time has been spent building up a relationship.

[1] or "investment", as even the otherwise legit ones are a gamble

> If someone purporting to be DHL contacted you out of the blue to say you needed to pay their fees in order to get this wonderful expensive package of yours, when you're not actually expecting a package, then that's clearly a scam, and you'd be daft as hell to pay

Well, yes, that is how these scams work - they spray out a huge number of such messages and rely on the chance that a small number of recipients will be too rushed[1], too naive[2] or simply too far on the decline[3] to take what we, the "clever lot", deem to be "the obvious precautions".

[1] life gets on top of all of us at some point

[2] the young and the trusting are a joy in life

[3] that fate awaits most of us

Re: "I am very smart"

Woooosh

Version 100 of the MIT Lisp Machine software recovered

Re: The Forgotten Fifth Generation

> First of all, there was a computer from the Japanese ICOT "Fifth Generation Computer Systems (FGCS)" project that was commercially sold...

Thanks for that info.

That particular footnote [4] was supposed to say "as far as I'm aware, updates most welcome" but you may have noticed I was having trouble keeping the footnotes in line.

> , but if you gave the expert systems more than a few hundred rules to follow in their decision trees they became painfully slow at returning answers.

Can you provide any citations for this? Serious question, not snark: in about 1985 we had an Expert System Shell applying simple Bayesian stats and I don't recall ever having to worry about it becoming painfully slow at calculating as you increased the rule set - it might start to ask more questions to help it traverse the data, but generally IIRC it organised the rules along the lines of a decision graph, walked from one side to the next, choosing the next node in the path based upon the data & weightings so far: a "wide but thin" set of k nodes would be solved faster (as in, it asked fewer questions before finding a sufficiently likely node that couldn't be disproved, aka "the answer") than a "narrow but thick" arrangement of the same number of nodes that arranged themselves in a different fashion.

But there are other ways of working an XPS (e.g. it need not be interactive with the User - ours was and so any calculation time was peanuts compared to waiting for them to type) and it is never too late to learn about them.

What helped kill off XPS (in the public eye) from our p.o.v. was mainly not showing a good enough cost/benefit: creating and validating rule sets is expensive and takes time away from your human expert. Simple XPS that could be created cheaply, such as an initial triage scheme, could be done as fast by the human operator once they'd run through the system a few times (i.e. the humans were trainable! Who knew?). Large rulesets (from our set of clients at least) were used too infrequently compared to the creation costs (and we didn't get into any of the big funded research studies). And some fields just learnt embarrassing lessons: like Stock Market trading. They can afford the set up costs, given the potential returns, right? But remember all those trials that showed that a monkey, an octopus or just throwing a die tended to give better returns on the market than the high-paid, so-called "expert traders"? And you *tried* to get an octopus to work alongside a Knowledge Engineer? Or vice versa, for that matter (the suckers! The horrible, horrible suckers!).

And back in the day there was something called "ethics" that controlled the use of computer models released into the wild: the XPS could show how it worked out a path to its answers, but if there wasn't a human around who could validate that when someone disagreed, the ethical liability was too high - and your human experts were not wild about being tied forever to that project, especially if they were planning to retire...

Luckily, all the current crop of LLMs and GANs etc are totally incapable of being queried for how they reached a result, so there are no ethical restrictions on just dumping them into the wild.

> more standardized, general purpose, and better-funded platforms to steamroll over them and crush them in the market.

True.

Re: The Forgotten Fifth Generation

[4] sigh.

Re: The Forgotten Fifth Generation

Quirks aren't necessarily bad, although they are things you have to be aware of in order to get the best out of whatever it is that is quirky.

For example, to make the best use of an old timey DOS PC, TurboPascal has the ability to do overlays - which you certainly didn't read about in the "Pascal User Manual and Report". You could quite easily learn all about the Fascinating World of Thrashing Disks using overlays, especially fun if you gave a hard drive and your Users have dual floppies. Not unique to TP but it made them so easy you could just sort of wander into using them.

Even just if you were, at the time, used to big multi-pass compilers that would eat your sources and spit out an error listing twice as long as your code, with every single missing semicolon mercilessly highlit. Switching over to TP's "halt on the first typo" did require a change in attitude. True, TP compiles really fast, but one of the ways it does that is to not be able to continue past that typo.

I tend to think of the various Turbo products as being pragmatic more than standards based, so you definitely needed to run through the manuals to find the odd bits that allow you to write code that'll give access to the PC's whole feature set.

Re: Good or just entertaining

> it's obvious that they're setting up a straw man as the method they don't like

Did you read the second URL, https://www.dreamsongs.com/WorseIsBetter.html ? Possibly not, as the article describes as an excerpt of the longer piece.

However, it is actually a fun meta-piece about the "Worse is Better" idea, where the author describes how, over time, he flips between both points of view, including writing (and having published in the same organ) articles where he argues for both sides (battling between his own name and a pseudonym)!

The Forgotten Fifth Generation

What can be seen as another attempt to take the path started by the LISP machines, the 1982 to 1992 (approx) Japanese "Fifth Generation Computing Machine" project, funded by MITI, is worth a mention in passing, at least. Even if its greatest legacy seems to be the stupid name[1] applied to database languages from 1983 onwards!

As with LISP machines, the intent was to create hardware that could run a language suitable for (among other things) creating small programs that can integrate together and work on top of a rich environment: as the article notes, an approach that Smalltalk also allows. The difference is that the language chosen, Prolog, was also deemed to be one that would work well across a multiprocessor environment, as the (putting it crudely[2]) search tree could be run in parallel along many branches at once, as well as working with a distributed set of "database clauses/statements"[3].

5G machines are less remembered than the LISP boxes (perhaps because they never[4] were actually sold, so few people ever saw them - or just because they didn't produce anything as memorable as the Space Cadet Keyboard!). Prolog has faded back into its niche, despite its moment of public glory (MicroProlog on the BBC[4] and Spectrum, even Borland had TurboProlog[4] for the PC).

[1] you know, calling them 4GLs, because the marketing twonks saw the slogan "Prolog, the Language of the Fifth Generation", thought it said "Prolog is the fifth generation of languages" (confusing s/w with h/w) and decided that the rash of languages for PC databases in the 80's had to be the generation behind... (Please, don't try to fling the Wikipedia article as a rebuttal!)

[2] very crudely, and let us ignore 'cut' for the sake of a brief comment!

[3] crude, I said crude!

[4] the beta ROMs on the BBC ruined my maze-solver for weeks, because we'd not noticed the "debugging aid" that was the top left character changing to show the garbage collector state, which ruined the Mode 7 graphics!

[5] TurboProlog was - quirky - compared to, well, every other Prolog; then again, it was released by Borland, so no-one was surprised.

FTC urged to freeze OpenAI's 'biased, deceptive' GPT-4

Re: It is singularity already

> It's not going to be creative because it was written not to be

That poses interesting questions about how to interpret the training of these models: what is there in the process of creating a statistical model that is explicitly saying "do not be creative"?

Re: It is singularity already

> Brains cannot be magic. They must be statistical machines

An old colleague used to point out that models of how the brain worked reflected whatever the current information technology was.

Plato described a wax tablet in our head.

bo111 thinks it must be ChatGPT.

Re: It is singularity already

> They are about any symbolic relations

Do you have any citations to support that?

The output from a LLM can be passed into a symbolic evaluator, but, so far, the only "support" for general symbolic processing I've seen has all been down to the same old statistical "if you see a mention of X then it is likely that the next bit mentions X-prime".

Given that they are well known to contradict themselves, if they are *supposed* to be proficient at say, symbolic logic, then I've got some old books on inference engines they may like to read (hint: when doing reductio ad absurdum you're not supposed to report the negation as the final conclusion!)

Uptime guarantees don't apply when you turn a machine off, then on again, to 'fix' it

Re: Not following procedures?

The advanced version of Victor Borge's routine for real punctuation enthusiasts:

https://www.youtube.com/watch?v=TIf3IfHCoiE

(aka "curses, couldn't find the video I'm looking for and the edit window is about to close)

Re: Not following procedures?

>> What sort of parents would name their child with enclosing quotes.

> As Rod explained in the story: internal politics.

So sad when families start to filibuster and demanding that voting lines be redrawn (he is only your step-grandfather!)

Ending up, at the christening, with the mother handing over a note reading: his name is "Rod", not "Roderick". "Rod"!

They were lucky the pastor couldn't pronounce the pling.

https://www.youtube.com/watch?v=Qf_TDuhk3No - Victor Borge, Phonetic Punctuation

Re: Documented End-User Procedures

> with literal checkboxes

So that's one checkbox for each of the times you type 'sync' before the 'shutdown -h'?

Scientists speak their brains: Please don’t call us boffins

Re: Let's save time

To answer a question from the end of that piece "Why not just refer to scientists as exactly that? Scientists.":

Bluntly, not all scientists - purely as "people who do science" - are boffins. As in all fields[1], there is a range of abilities, even if the curve is offset from the wider population. Your boffins are the best of the best.

[1] I'm still allowed to use that derogatory term here, yes?

Re: Hmmm

"Abigail's Petri"

Re: Icon alert!

Indeed; if we were discussing Mad Science then Rotwang fits the bill, but a proper Boffin will be reaching wildly for the mug of tea[1] whilst in his slippers.

[1] which the long-suffering assistant nudges into his hand in the hopes of avoiding Another Incident: whilst it did lead to a breakthrough, the damage done to the wainscotting will never be fully repaired.

Re: Deter people from studying in the field?

> Except if you could tap into ZPE, how would you patent this

Surely "Star Gate: Atlantis" is prior art, rendering such a patent invalid?

Re: Deter people from studying in the field?

> There are six types, known as flavors, of quarks: up, down, charm, strange, top, and bottom

No, no, you can't use "top" and "bottom", they have far too many connotations and will upset people.

We'll just have to go back to the old alternative names for the t and b quarks, "truth"[1] and "beauty".

[1] As t is the last on the list to be found, "The Search for Truth" was a much better title than using "Top"!

Re: Hmmm

Hmm, the problem there is that the "ette" ending has the primary meaning of "small" and the feminine is a secondary meaning - compare with the masculine ending "et".

So "boffinette" and "boffinet" would be the gendered words for, say, anyone pre-doctorate.

(In case you were wondering, a "(drum) majorette" is a diminutive, female, version of a drum major: the latter manoeuvres a large baton, approximately[1] 2m, to keep the marching and in time and directed - ref. the end credits of "Thunderbirds Are Go!" - whilst the former flings about a small baton. Guess what I think the word "bachelorette" means!)

[1] approximately!

The most bizarre online replacement items in your delivered shopping?

Re: They don't make it easy.

> If substitutions were allowed in software development, who knows how the UI would turn out?

We already know: ribbons in place of menus!

Re: Sellotape instead of loo paper

You mean: was the tape brown before it was unpacked?

Utah outlaws kids' social media addiction, sets digital curfew

Re: Dumbass Legislators of Red States

> Kids slip on and off of the darknet

Do you know anything about hackers?

Can you jam with the console cowboys in cyberspace?

Never experienced the new wave? Next wave? Dream wave? OR cyberpunk?

Re: While I do not agree with 99% of...

> there are no caveats to protect the children from abusive parents.

If parents are using social media to abuse their children why would they need extra caveats compared to anyone else using the same social media to abuse the same children? Abuse is abuse, no matter the perpetrator.

Or are you thinking that legislation over social media must also carry some riders to extend its scope beyond social media?

Will they bother analysing the causation?

> the introduction of Facebook led to increased utilization of mental healthcare services.

This could be for (at least) two distinctly different reasons:

1) They learnt about the services on FB and were encouraged to seek them out, directly by concerned friends or indirectly by reading how others had gone and found the services useful, their publicising the issues helping to remove the stigma.

2) They encountered a torrent of crap on FB until they cracked and stumbled or were even dragged in.[1]

But what are the chances that such subtleties[2] will be ignored when trying to control Social Media? And then the good effect encouraged whilst the latter is (somehow) reduced?

[1] Sadly, I'm thinking this is the more likely scenario.

[2] Subtle! Those two examples?! I have no faith that legislators could understand the effects of, say, shyer college students seeing "**everyone* else" going wild without them, or that those posters are themselves "faking it 'til you make it".[3]

[3] Bugger, how I wish I could come up with a "more subtle positive" and not just two more negatives.

How Arm aims to squeeze device makers for cash rather than pocket pennies for cores

Re: Options for Much Silliness

At least that would prompt companies to write optimised code and get rid of some bloat.

Errors logged as 'nut loose on the keyboard' were – ahem – not a hardware problem

Re: Installed Remedy

Not yet been subject to Remedy or Service Now.

> Internally we used Jira which is much nicer

OMG! What must life be like for Remedy or Service Now users!

Silly devs, Bugzilla is too fruity for you, it's just for Managers

Join a dev team that is not exactly up to speed on the latest techniques and, after introducing Make to replace the compilation batch files, asked if I could start a Bugzilla server for devs to log into. Quite happy to start it locally on my PC if that helps kick-start the process.

Some years (!) later, we finally get Bugzilla approved and installed (at least it was on a server in IT, so that bit was good) and devs start to use it. Including noting that you can mark an entry as an "Enhancement" - not actually a bug or problem, but a good idea either from a customer or even devs themselves. We trundle happily along, including BZ references in commits and even in release notes. As you spot links between BZ entries you link them together and add notes "turns out, this if really easy if...". This is an old-style program, we do new releases at varied times, to match bug fixes or new functionality for new contracts, rather than forced releases every three months for now good reason.

Time passes and even Project Management learn how to enter bugs. Then PMs start dumping out stats and track how long BZ entries are left open. One day, management gets a Dashboard web page and the PMs stats are proudly offered up.

But they don't bother filtering out Enhancements or any filters at all. "It is called Bugzilla, it can only be used to hold bugs" and "We don't need devs to help with this, we know about Dashboards and our web guy knows how to put raw MySQL data into an IIS page".

One day, my BZ admin screen shows a hundred or more entries have been closed and a huge number of refs between others removed. Our direct manager has pruned out all the items that showed up as "too old, must be closed", including all Enhancements. Someone has read enough of the manual to be dangerous, got admin rights and removed everything but bugs from the input screens.

A new customer, new project, new requirements: all changes get entered as "bugs", we aren't allowed to link them to old entries that discussed how to do any of the items we'd already discussed as Enhancements.

The Dashboard bug graph peaks, blood pressures follow, the older devs grind their teeth to dust...

New contract arrives

Microsoft scrambles to fix Windows 11 'aCropalypse' privacy-battering bug

Re: Win11 TPM and Secure boot, makes zero difference to poorly written signed code.

With one mighty bound...

Re: I wonder

Ignoring the fact that the Android issue is that the OS changed its behaviour and stopped truncating the files, breaking the programs.

> thus foot gunning all the clients.

WRT the Android bug, I've been a bit weirded out that (so far) everyone I've seen discussing it is *only* talking about how that can affect images, when it can clearly affect *any* data where the new data is shorter than the old. Just as long as there is an EOF mark or "chunk size" field which tells the normal file loader/parser/displayer to stop. Even down to plain text, if the EOF character(s) are being correctly honoured.

Why expect the PDF converter to do more than Word did?

> I think there must be a misconception amongst devs that "crop" really means "frame"

Programs like Word will clearly only set a view onto any image that you drop into a document, keeping the entirety of the original image within the document. Otherwise, you would not be able to go back into the document and still be able to change your mind about how to present the image in document (e.g. decide you want to alter the crop to match some change in the text).

When converting the document to a PDF, the simplest (as in, least likely to go wrong) option is just to do the dame as Word: set up a view onto the original data. And not attempt to scan the document to see if it contains multiple copies of the same image and try to perform any extra optimisations that Word hasn't bothered to do.

Unless there an explicit claim somewhere within the documentation (or sales blurb) that the PDF converter is going to space-optimise the result (as some products do) then the PDF converter has done exactly what it promised to do, no more and no less.

B-List celebs including Lindsay Lohan fined after crypto shill probe

This is not the L.O.H.A.N. you are looking for

For an odd moment[1] there I wondered if the SPB had decided to launch an NFT[2] instead of a paper plane.

[1] Another too-early-in-the-morning-for-the-Register moment

[2] Nominally Flightworthy Transport?

Amazon to shutter Digital Photography Review

If you will forgive me copying from my own post of yesterday

Ever since happily stumbling across it, I've not bought anything photography related until I've had a chance to check DPReview for the specs & reviews. Yes, there are other photography sites, but I appreciate the consistency and completeness of their kit database. Plus the fact that they don't take themselves too seriously.

So, dpreview.com is to be shuttered in April, as Amazon wants to concentrate on Project Kuiper, Prime Air and Smart Homes. All of which will, of course, encourage me to spend more money with Amazon than I ever spend on photography. Not. Project Kuiper: Space trash. Prime Air: Sky trash. Smart Homes: Ground trash.

Pah.

Journalist hurt by exploding USB bomb drive

Random USB devices can usually be handled safely[1]

and as, of course, journalists are going to get unknown USB sticks from sources unknown and need to access the contents, I'd've hoped that news outlets would provide the necessary basics and training to do so[2].

Sending exploding devices can only be condemned, of course, but hopefully the journalists won't be cowed by such tactics and will just respond by taking obvious extra precautions - a long extension lead and sand bucket for a start.

[1] as you all well know: use a cheap copier machine, preferably one without a complex OS, always have autorun disabled all the time, virus scan after copying to trusted medium but before opening, disable macros in Excel etc; hardly earth-shaking stuff.

[2] ever the optimist, but come on, this should be mandatory basic stuff, like fire drills: it's negligent not to provide it.

German political parties accused of microtargeting voters on Facebook

Re: Meta

> If your facebook 'friends' are not your real friends...

He is building up a fake identity, a "legend" if you will, so that When The Revolution Comes he can disappear, like the Oozlum bird.

Are you ready to go all-in, head-first, on a laptop? ASUS's Zenbook Pro 16X asks for that commitment

Re: keyboard that tilts upwards to put it at a more comfortable angle for typing

Hmm; on the other side of the table, I always pop down the feet at the back of the keyboard to raise it up, as that feels much more comfortable to me.

Maybe it is a difference between using flat, unsculpted, low-travel laptop keyboards[1] versus big chunky "mechanical" keyboards with sculpted keys?

Either way, different keystrokes for different folks.

[1] sometimes I get jealous of those who are comfortable using those keyboards, as they appear to actually enjoy using laptops and can happily take their stuff anywhere: I've tried coding in grassy fields or the hotel lounge and can only do it in extremis.

SpaceX tries to de-orbit Amazon's request for a satellite broadband shortcut

Re: Hugo Drax Wannabees....

On second thoughts, Let's Go Drax!

Pew, pew, pew.

Re: Hugo Drax Wannabees....

I've been hoping for something more like You Only Live Twice: to the accompaniment of a thundering good John Barry style theme, one mad billionaire's satellites swallows up another's.

Whilst Musk's lot would be sticking to the tried and true "bring the captured back down to my volcano^^^^^^^barge-based lair" script, Amazon will have to follow the Body Snatchers approach, suborning the SpaceX satellites and turning them into its Own Kind. Hmm, they want to share the same frequencies as Starlink, eh? So the zombie Starlink sats will have the correct radio kit to join in with Kuiper? Suspicions are growing...

Vessels claiming to be Chinese warships are messing with passenger planes

Re: A better solution

> scorched earth

So THAT'S what you guys make that stuff from!

Software-controlled food tech: 3D printed pipe-dream, or fatal stack instability?

Glad to see they've been reading old Make magazines

and blogs from the past ten, twenty years. It may be a breakthrough for them, but...

These things have been toyed with for ages by enthusiasts, printing with anything that can fed through an extruder and/or cut by a CNC bed with a novel tool head.

Not to mention the robot mixologists (vital during the development stages, when the laser-cut gingerbread catches fire or the pancake batter drawing bot starts serving up obscene shapes).