Posts by Alan Brown

Re: This is an old, old problem

"HR normally have to have it"

HR have to have a number.

070 follow me numbers cost a fiver a year and cost £1.50 per minute to call. You won't see any revenue from that but it's a powerful disincentive for callers.

Re: This is an old, old problem

"A few years ago I even had a complaint that my personal phone was switched off."

'How did you get hold of my personal mobile number? I didn't give it to the company....'

Re: First in to work....

" So I rang the "engineer"'s office and spoke to the service manager there."

Did you suggest to the client that they trespass the "engineer" in question so that the AC company would have to send out someone competent?

Re: The quiet hero almost never gets the beer.

The IBM of olde had a LOT of pride in their kit, their company, their customer support and encouraging people who were taking steps into computing hardware to discover the products - various folk in the company had my back when I had to take steps to kick the ISP part out of one of the larger IRC networks for a prolonged period because of chronic network abuse coming out of Asia (cyber warfare and script kiddies are nothing new). It was the support of those staffers which enabled those bans to stick AND to force IBM to make policy changes about the way they owned handling of abuse originating inside their networks instead of disowning what their customers were up to.

The attitude coming from those parts of the company made it clear that the days of the "IBM of olde" were numbered and whilst Marketing and Sales were eventually spanked that time (quite hard - at least one of the managers involved was "removed", only to show up at another ISP shortly afterwards) it wasn't long before the tail finally went into full wag the dog mode and the company lost a lot of very good people.

Re: Totally different industry...

By the time it got back to France it was probably ready to blow the bloody doors off without any external assistance.

and I'm betting that there would be debate about even _allowing_ it back on the ferry or chunnel.

" The air con died during a scorching summer period 2 years ago, quickly raising room temperature to high 50's. "

Won't happen in my server room.

Mainly because there's a crowbar dropped on the power rails if the room temperature goes over 35C

Most of the kit is now setup to respond monitor its IPMIs and if the inlet temperature goes high, will power down before that point anyway (dumb little cron jobs)

It's easier to recover from a power outage than dealing with cascades of failures resulting from a mega heat excursion.

Re: Something similar happened to me...

"Right up until the equipment changes (gets added to, or rearranged for prettyness without regard to airflow) and the air conditioning gets left out of the considerations."

Which is why ass-clenchingly high paranoiac levels of power and environmental monitoring are ALSO required.

Get it on the way in, because one the kit's up and running you'll spend decades arguing to get that level of monitoring to know what's likely to overload where and when - until about 15 minutes after the shit hits the fan in a massive way, at which point you'll get the blame for not having emphasised the scale of the problem being of such criticality to the business. (You can turn this into a "How much money do you need to fix this?" and a "How much money do we need to avoid recurrences?" moment)

Re: Torvalds declared: "Don't use ZFS. It's that simple."

"BtrFS promises similar, but to my limited knowledge is not nearly as good as ZFS "

After several total loss incidents during testing of BtrFS that ZFS simply shrugged off, I think you're highly optimistic.

Everyone who cared about data integrity jumped ship in the development arena years ago, so the prospects of BtrFS improving are slim to negligible. The better mousetrap already exists.

Re: Torvalds declared: "Don't use ZFS. It's that simple."

"Nouveau drivers are for masochists"

There's a simple answer as explained by several of the gurus on #nouveau in the usual open source IRC network:

"Don't buy Nvidia products and expect them to work on Linux. Even if the drivers support your device NOW that can stop at any time without warning. Nouveau is a best-effort project and Nvidia have consistently been less than helpful whilst providing constantly moving targets for their largely undocumented APIs. Nvidia simply don't care about Linux support as it's such a tiny area of their marketplace"

In short - and this is the advice from the Nouveau team themselves - if you want it to simply work, buy AMD graphics cards/GPUs

"It will be Ubuntu which may prod Larry's lawyers into action as they're looking to include it in their installer."

If the installer downloads the modules or compiles the sources to link into the kernel on the fly then that's perfectly legal and acceptable. Oracle attempting to stomp on it will come off second best in such a stylish.

Re: The problem is not Oracle (for once)

"GPL demands code to be 'eternally open', no additional requirements (like a copyright)."

Bzzzt. Incorrect.

GPL is fully copyright aware. The license conditions are such that if breached then the license is void and what results is a simple copyright violation - with inherent criminal penalties in the USA if there's more than $50k involved.

This is why the very few cases challenging GPL have collapsed. You can comply with the license by publishing the derived source code, or you can face potential jail time for criminal copyright infringements. Once judges point this out the arguments from lawyers dry up rapidly...

This is like the case in the UK where water companies tried to claim ownership of all water falling from the sky and prosecuted people collecting rainwater. When walked through "their" water (and associated liabilities) causing property damage through flooding etc, the prosecution lawyers asked for a hasty adjournment, then dropped all cases in less than an hour - they'd realised they might prevail on water ownership but in doing so, open themselves up to hundreds of billions of dollars of liabilities and well-heeled insurance companies claiming their repair costs back.... (it's called a pyhrric victory)

Re: The problem is not Oracle (for once)

GPL existed long before CDDL

Sun released other code under BSD or GPL licenses

It's been carefully and thoroughly explained by many people many times - several of them with firsthahd knowledge of the events in question - why CDDL is deliverately incompatible with GPL and why that was done.

As for "Blame Oracle" - the very first thing they did after acquiring Sun was to cancel ALL the open source licensing on Solaris-related products. OpenZFS (and opensolaris) is forked from the LAST available CDDL ZFS version - more than a decade past.

Oracle have _no interest whatsoever_ in reopening ZFS under any kind of license, let alone changing legacy licensing. It's been clear for a very long time they'd prefer to invalidate CDDL without a replacement, but thankfully they can't (yet) do that retroactively until they grease the right palms in various legislatures.

Re: The problem is not Oracle (for once)

The BSD vs GPL issue is simple.

Under the BSD license you can take my code, roll it into your proprietary product, hide the source or implement secondary restrictive licenses and sue ME for copyright/licensing breaches on code I helped write.

Under GPL you must make derived source code available when distributing the product or the license is invalid and your code is ow in breach of MY copyright.

What BSD makes it "freer" to do is steal code and falsely claim ownership of if. It's EXACTLY that experience which motivated many people to switch to GPL - including myself. There are a number of commercial closed-source solutions floating around on the net containing large unattributed chunks of my work.

This is also the reason Microsoft used to love BSD licensing and rubbish GPL. It's an effective license to steal from the commons without giving anything back and naively relied on basic honesty that's lacking in most businesses. I've always suspected ethics classes were a front for surgical removal of any that might remain, before cauterisation of the stump.

Re: Hypocritical

FreeBSD's network stack isn't quite as good as Linux' at high speeds/loads

There are a couple of other reasons why people favour ZFS on Linux over FreeBSD or openslowaris, but it's not all just about personal OS preferences.

As for CDDL - Linus is absolutely correct to avoid this mess until the license is changed. CDDL was specifically written from the ground up by Sun lawyers to be incompatible with any form of GPL because Linux was seen as a clear and present danger to Solaris. BSD blindsided them.

The current setup is a little messy but as long as ZFS and Linux are kept separated until fused on the enduser box AND as long as hardware isn't traded with ZFS-on-Linux installed then it falls within the licensing limits. (Ie, they can only be combined on the enduser system AFTER the enduser has taken physical and financial possession of the system)

Re: Hypocritical

The law doesnt differentiate libel or slander - it's all defamation.

And contrary to popular belief even in advanced civilised societies, truth is NOT an absolute defence against defamation. (Posting about Fred's expunged criminal record or that Karen was a prostitute 30 years ago being two obvious examples upheld in UK and USA courts despite the latter having plenty of evidence to back it (Karen left that behind long ago and is a respectable lady now) the former being part of warnings related to "Fred" engaging in the same behaviour that got him convicted 10 years ago)

In less enlightened countries *ahem*France*ahem* you can even be found guilty of defamation of a public figure by way of posting a particularly robust opinion or criticism.

Which all means this law is mostly used by those with deep pockets to gag those with less deep ones and only occasionally for the purposes for which it's apparently intended - and even then if the defamer has deep pockets the defamee seldom prevails.

Re: Hypocritical

That was a different company. The SCO name changed hands a few times along the way and the SCO that started the litigation was not the Santa Cruz Operations that created SCO - which remained part of Novell at the time.

In a further twist of irony, it turned out that SCO really dodnt have standing to litigate over alleged SCO violations as they didnt have full copyright licensed to them by Novell - and Novell weren't at all happy about SCO's antics or claims of ownership of SCO

Re: Cancelled my BT account

"But something has to change for us country folk.!"

It's about to.

Monopoly telcos worldwide are cacking themselves at the realisation that Elon's Skynet system will impose a HARD upper limit on what they can get away with charging (or filtering) and it's interesting to see the funding trails for some of the "astronomy action groups" that trace back to the usual astroturfing organisations.

Heaven forbid that a midwest USA monopolist telco that's been greasing the palms of the PUC for the last 30 years suddenly be exposed to actual REAL competition, Not long after they've managed to drive everyone else out of business.

Re: Cancelled my BT account

" We've wasted nearly £1BN on rural broadband so far with almost nothing to show for it."

New Zealand showed how it can be done. Make ANY further money contingent on the lines and the services being entirely cleaved from each other - separate ownership, stocks, boards and directors.

Having BT own the infrastructure just perpetuates the abuse in ways that New Zealand's Ministry of Commerce enumerated in a lengthy manner before rejecting Telecom New Zealand's attempts to sell the "BT model" there.

Re: Emergency calls

"That to me is a 50v exchange line, backed with at least 48hrs of reserve capacity."

I've lived in areas where phone battery was two #6 dry cells powering the microphone. They had a lifespan of 20 years or so. Central battery was mostly invented so that telcos didn't have to send out battery techs to do swapouts on a rolling basis, but helped a lot with decadic (pulse) signalling later on.

Re: one major problem that Ofcom is deliberately ignoring

"In theory, if you pump enough energy down the fibre cable, you could power things on the other end"

Right now even 80km fibre transmitters only put out about 10-20mW and those are regarded as dangerous as all hell. Most shorthaul optics is around 1mW and it can still put holes in your retina if you're not care, thanks to the small spot size.

If you want to put out enough light to power things at the other end the power density required would probably MELT the fibre.

So in theory, you could do it, in practice you're getting toward the arena of weapons-grade lasers.

Re: Spare a copper?

"Backup batteries at the cell sites are not maintained or not existent.."

If Ofcom aren't making this part of the operational requirement, then someone should be asking questions about why not

Re: reliability

No, it's steel, which is almost worthless in the quantities found in cables.

Re: Ridiculous

It seems to me that any software manufacturer could simply move out of the country.

there, FTFY

Re: And the consequenques fo failing to act?

And yet, the biggest threat I see to Iranians I know is arbitrary detention without trial BY THE IRANIANS(*) simply for having contact with the west.

(*) Their security services and revolutionary guard are hyper-paranoid

In other words, If you're Iranian and you leave the country, you're in double jeopardy

Re: My only memorable incident

" No problem! I advised her to keep the keyboard clear, in a pleasant, calm tone, it could have happened to anyone. I was later asked about the incident, behind closed doors..."

I was relatively clear and pleasant when I asked the user why she'd deliberately disabled the antivirus package that was doing its nut when she tried to open the virus-infested email attachment that ended up trashing not only her computer but the rest of the (thankfully isolated) lan it was on ("It might be important") and why she thought it was a good idea to do so without bothering to consult with IT, and end up tieing up 2 of our staff for 3 days - not just once or twice, but THREE times.

She turned around and filed a formal complaint about being lectured as if she was a naughty five year old child...

Apparently my response that said user would be better off given an etch-a-sketch and some formal written warnings about unauthorised interference with IT equipment caused much hilarity as well as hitting a few raw nerves. Not long afterwards we were authorised to lock down the systems involved so she _couldn't_ repeat the offence and she rather pointedly avoided me for the rest of her employment (If I'd had my way, pulling that 3 times - especially as she'd been warned by IT the first 2 times - should have been a sacking offence. These days each event would have resulted in a major GDPR breach report+investigation)

Re: Floppy drives

"like other commentards here, I have personally used magnets over the years and can't guarantee the loss of information."

I use them all the time - to attach USB sticks to things.

The day one affects the contents, I'll eat the stick.

Re: Paper tape corruption

"a bored halfwit poking the occasional random extra hole into the tape "

There's a story of punch card stacks routinely being corrupted when exported to france - a QC checker was eventually assigned to follow the entire shipment to see what was going on.

French customs would open the shipment, pull a sheaf cards from the deck ("a sample"), look at them and throw them away. No amount of explaining that the entire deck was "ONE DOCUMENT" and what they were doing was the equivalent of ripping pages at random from a book would sink in.

A marker pen trace laid across the top edge of the deck made such missing bits more glaringly obvious though.

Re: Stray magnetic fields...

"I started saving programs onto *two* cassettes, one left in the lab and one taken home with me."

Experience with 1980s computer disk drives showed pretty poor reliability, even without the added spice of NMR equipment nearby

Re: Erasing hard-drives...

If you need to deguass to that level, a nine inch nail and sledgehammer punching it through the platters is cheaper, faster and FAR more satisfying.

Followed by a nice baking.

As far as reading with an atomic force microscope: Peter Gutmann's own followups to his seminal paper is worth reading:

https://www.cs.auckland.ac.nz/~pgut001/pubs/secure_del.html - read the epilogue(s)

TL;DR: Those tests were on 10-20MB MFM stepper motor drives, attempts to repeat on 100MB voicecoil drives failed, higher density drives are unlikely to work either