Posts by Doctor Syntax

Sweet 16 and making mistakes: More of the computing industry's biggest fails

Re: Honourable mention

The first Unix was certainly written for the PDP7 but was that in a high level language of any sort as opposed to assembler - or even raw machine code?

"it was clear that what buyers really wanted was not multiple apps at once, but the point-and-click ease of a GUI."

Or possibly they wanted reliable storage. Sinclair had got away with cheaping out on build quality with the Spectrum and obviously thought they could do that again.

DataVita declares sovereignty with 'National Cloud' for UK

If I understood this correctly it runs on an HPE facility. HPE is a US company. As such it's subject to the usual US pressures such as the CLOUD Act. How does that amount to "sovereignty"?

TikTok isn't protected by Section 230 in 10-year-old’s ‘blackout challenge’ death

Re: At last, some common sense

"Ah, good'ol autoerotic asphyxiation."

Nothing good about it when it eventually ends up with the family visiting the coroner's court. It's not a situation that I had to investigate myself but I've had colleagues who did.

In this case it seems the intentional abduction is the closer analogy.

Alternative view: Tik Tok and the parents should be liable. The two are not mutually exclusive.

Re: At last, some common sense

"So they select content about asphyxiation, and users actually read it and/or try what is suggested. Who'd have thought it?"

Anyone who had experience of dealing with sudden deaths. Distinguishing between what's generally termed sexual asphyxia and suicide was a problem well before social media, the web or even the internet existed.

Re: Excuse me whilst I......

I take it you're from the US.

The Canadian air line case made it the responsibility of whoever deployed the AI. I don't know whether that was at a court high enough to set a precedent or whether Canadian precedents would be followed by a US court. In any event it seems a reasonable line to take.

'Uncertainty' drives LinkedIn to migrate from CentOS to Azure Linux

Re: it's not uncertainty

I agree. I also think that's what the OP said.

More advanced != better

Re: Why so complicated?

They still need to keep the data in sync. A user in the UK would need to see the same things as one in the US as one in India as one in Australia even if they're accessing different data centres..

Given that there are other UIs for Linux on the one hand, and that the Windows UI could them be installed on other Linux distros it would open the door for mass desertion by their user base.

"MS and Oracle teaming up would indeed be interesting"

Interesting as in the Chinese curse. It would be time to move to a BSD.

As to RH making SystemD & Gnome proprietary, having the rest of the community simply turn away would be better than forking them.

Re: Why so complicated?

Running at that scale is a non-trivial problem. They're running from multiple centres but still have to keep the data in sync.

Then there's risk aversion. Because of that production distros such as Centos tend to be trailing edge in terms of release versions of different components but with security updates back-ported. They're going to have to build and, also because of risk-aversion, thoroughly test all their custom stuff onto what may also be a distro similarly customised for extremely long term support but with a more recent base.

"If you want to improve your company product make your company actually use it in real world conditions"

This makes me wonder about their patch teams. Do they really put up with their own product? Or maybe they do and have no experience of how much better it could be. Perhaps they should be made to run something like Debian as their daily drivers for a couple of calendar months and then switch back to be faced with the accumulated junk they'll have to catch up on.

Brit teachers are getting AI sidekicks to help with marking and lesson plans

At least none of the chatbots told you there were 3.147 or anything in raspberry.

HP secures $50M CHIPS Act boost to adapt inkjet tech for life sciences

Re: Wanna Equity!

" the terrible, dreadful, etc. etc. (so we're told) Labour government of the 1970s"

I don't need to be told. I remember it.

Are they hoping the reputation they've acquired in recent years hasn't preceded them into the lab market?

Or are they hoping there are still people who remember the HP of old as being one of the great makers of lab equipment? But that was spun off into Agilent end other fragments long ago.

Microsoft hosts a security summit but no press, public allowed

If you've nothing to hide...

EU tries to pin down China on definition of 'important data'

"In June of last year, the CAC instituted even tougher rules that required regulatory intervention when sending data abroad. Companies that move data describing over 100,000 individuals or handling information of over one million people were required to conduct risk assessments and file declarations."

Tougher? That sounds remarkably lax.

Where the computer industry went wrong – the early hits

Re: The big mistake

"and the network effect reinforced that advantage"

Not the only time that that mechanism has left computing stuck at the level of the barely adequate.

Re: Two significant characters?

Except possibly Forth and APL.

The engineers seem to have lived in perpetual conflict with Jobs once he took over the product. They wanted to make it expandable and he didn't.

Re: Flogging a dead horse

It leaves customers in both the low and middle range feeling ripped off. The customers for the top range are, of course, being ripped off even more if you disregard flauntability as one of their criteria.

A third critical difference was the education market support for the Beeb. Possibly the last sensible computing decision by any HMG of any colour.

Re: Two significant characters?

I had a colleague who wrote BASIC programs in FORTRAN! 2 letter variable names, spaghetti and all.

Re: C128 the Hero!

QUB had an instruction lab running Apple ][ clones. I think they were actually made under licence but can't remember who by. They were running UCSD Pascal.

Re: C128 the Hero!

"I think I only ever saw real TRS-80s of any kind in Tandy shops."

My boss had one. I built a joystick interface for it based on one of Steve Ciarcia's articles in Byte. If memory serves that involved making a PCB with a Dalo (resist) pen and etching with ferric chloride.

Years later I saw a TRS80 on a market stall and bought it simply because it was a piece of history. It's in the garage somewhere - I don't even know if it was working when i bought it. By now it would probably need re-capping if I wanted to try it.

My local Tandy shop in NI had a range of non-Tandy IBM compatibles. They had a case design I've never seen elsewhere in that the lid hinged up from the back, no faffing about with screws or whatever. They were ideal for playing with using for insrumentation in the lab.

Re: The big mistake

"The blame should probably be shared between IBM and the clone manufacturers."

This is much later than the period of he article.

The point that everyone seems to be missing here is that when the first 8-bitters emerged there was no existing S/W except what was in the ROM. Somebody had to write it. The "vast software code base" was built up over time once there was something to write on and write for.

Hardware manufacturers made hardware. Software was an after-market product. And if the choice is between a new, more powerful video chip that's not compatible with the older one and he older one that's compatible with the older software what hardware manufacturer is going to put up with the pointing and laughing that would come with stagnating specs? They'd expect the after-market to catch up.

Would it always have been feasible to maintain backward compatibility whilst introducing new H/W features?

Re: Water under the bridge

The 8-bit era was for computing what the Pre-Cambrian was for animal life - a huge variety of forms created and let's see what survives.

Re: The big mistake

Apart from what was built in on the ROM there was no S/W and that might be no more than a BIOS. Microsoft BASIC was developed on PDP. The first time they say any part of it execute on an 8080 was when they took the paper-tape of the first completed version to Altair on paper tape.

Re: Colour and the PET

I'm pretty sure the one I saw in the Met lab was green. It's possible that production fitted whatever CRTs were available at the time.

The key to understanding the early 8-bit computers is that there was no established market for computers at that sort of price. None of the manufacturers would have had a clear idea of who would have wanted to buy them and it's a mistake to imagine that they were all aimed at solely at games although that might have been many vultures' and commentards' first experience of them. Once one realises that what one might have regarded as a gaming machine was conceived as having a much wider market the availability of CP?M is less surprising.

Essentially the arrival of the 8-bit processor had enabled the creation of machines whose possibilities were open-ended, vastly so from the perspective of the day. OTOH, not having any established market the makers had to put them out there and hope that they got the mixture of facilities right for some people to find uses for them.

They may well have envisaged businesses developing S/W for their own use with some of them then selling the S/W to others which, of course, enabled H/W sales. That certainly happened. I remember one event in the series the Beeb made at the time. A little old (so she appeared to me at the time) lady running a sweet shop had bought herself some machine - I can't remember what - and was writing her own programs. She was talking to the interviewer whilst doing something at the keyboard when there was a sudden, angry snap of her fingers as she'd obviously discovered something wrong; a sudden non-verbal communication, instantly understood and remembered over the decades.

Others, like myself, in mid-career in science or academia had had experience on mainframes, and subsequently looked enviously at the likes of PDP-8s, well above our budgets. We latched onto them. I had been looking at the possibility of some dedicated gadget to interface a microspectrophotometer to a tape punch with the hope of getting the QUB 1904 to process the tapes for me when a fire destroyed our wing of the lab and by the time we'd got up and running ot became feasible to set the whole thing up with S-800 kit. I know the Met forensic lab was looking at the same thing with a PET (it should be remembered that Commodore had an expansion bus that was based on the HP-IB). I heard, via an archaeologist friend in the Ulster Museum, that someone in the QUB Psychology Dept. had a machine (?Sol) which was a console format with the basic computer on a motherboard but with a few S-100 slots for expansion. From today's perspective it's probably impossible to imagine just how exciting it was for someone used to low budgets to suddenly have these possibilities open up.

Colour and the PET

"luxuries that would have been prohibitively expensive in the 1970s, like color and sound."

Of course it had colour: green! I can't remember whether it had a beep but it probably did.

I once saw a number of the later models on sale at Henry's in Edgeware Road. Maybe I should have bought one but getting it home might have been a bit of a problem.

The real mistake of the early PET was the keyboard. Why - just why?

Enterprise SAP users split between on-prem and cloud as migration challenges loom

"Our new innovations"

Does this mean that can make old innovations?

Broadcom boss Hock Tan says public cloud gave IT departments PTSD

"Broadcom boss Hock Tan says public cloud gave IT departments PTSD" and he thinks that's his job now.

Tenstorrent's Blackhole chips boast 768 RISC-V cores and almost as many FLOPS

"Blackhole chips" sounds like a hardware implementation of /dev/null

Woman uses AirTags to nab alleged parcel-pinching scum

Re: "police declined to pursue the matter"

"Leave your bike next to one that's worth more"

But once the one that's worth more is stolen you're becomes the target.

A last look at the Living Computers museum before collection heads to auction

As a member of a local history group we have the other side of this problem. We are occasionally given collections of documents but have no premises in which to store them. What to do? I'm gradually moving them towards a hosted NextCloud instance to make them accessible to members but that only lasts as long as there are funds to pay for hosting and it still doesn't help house the originals. Archives are not necessarily interested in them.

Re: Museums...

"unfortunately they didn't have any clear loan agreements with conditions for returning the items."

This is critical. Loaning an item is safer than donating it in the event that the museum may close - and even publicly-owned museums, or those set up as trusts can run into hard times and be closed. But it needs firm documentation.

Zuckerberg says Biden administration pressured Meta to police COVID posts

Isn't that strictly limited to forbidding Congress from passing laws affecting free speech?

Re: "I'm not trying to play politics or foment controversy"

I think the usefulness of "performative", if it ever had any, has already expired. It just means "something I disagree with".

Microsoft Bing Copilot accuses reporter of crimes he covered

It depends on what constitutes "as best it can". Search engines have never been good at that. Adding a fact-mangling engine is only going to make them worse.

The better response would be along the lines of "He has reported $Cases. We originally stated that he had committed these offences. That statement was a mistake on our part and is hereby retracted."

The harm lies in the original statement. Setting that right is important. Issuing a correct version on its own does no do that, neither does ducking the subject entirely.

Given that it seems likely to have been a generic error fixing it one case at a time isn't going to help.

The Windows Control Panel joins the ranks of the undead

"You didn't like it when we said 'deprecated' to describe what we're doing so we'll stop saying that."

Re: You said “command line”

The Unix shell was conceived as a means of separating the low-level OS user interface from the kernel. As the only option at the time was a character-based interface using teletypes it was implemented as commands issued as lines of text, hence a command line.

Because os the layered nature of Unix it was possible to replace one shell with another so although you might be thinking of the Bourne shell or, possibly more likely, bash there were many others, ksh and csh being a couple I've used. Although, as you say, they have become programmable (and also implement some commands as built-ins) the essence has always been invoking other programs and providing parameters to them. If one wanted to be unkind (which I wouldn't) it would even be possible to describe that programmability as bloat.

With the arrival of GUIs we new have GUI-based shells. But if it's a text based shell interpreting lines of text as commands, then it's a command-line shell. What else could it be?

The elusive dream of cloud portability: Why migrating workloads isn't so simple

Re: And I've been saying this all along

OK, Microsoft has your balls in a vice. What's new?

The future of AI/ML depends on the reality of today – and it's not pretty

Re: Good for fun

"useful examples without having to scour StackExchange."

It's just that the AI tool has scoured StackExchange for you. But has it taken not of the ratings of the solutions offered?