Posts by keithpeter 2067 publicly visible posts • joined 14 Jul 2007
Worth mentioning that OpenBSD 7.1 includes a change in the apmd power management software so that when plugged into the mains, a laptop will tend to run at maximum processor speed. See
https://dataswamp.org/~solene/2022-04-21-openbsd-71-fan-noise-temperature.html
for references and a user space daemon to provide sane defaults.
The multi-slice default automatic disk partitioning is apparently something to do with write XOR execute permissions on various slices, see
https://undeadly.org/cgi?action=article&sid=20160527203200
I'm always impressed by the documentation *within* an OpenBSD install. The FAQ and man pages for the base system are also available from openbsd.org. Reading
https://www.openbsd.org/faq/faq1.html#OtherUnixes
might help with the initial orientation.
The 'learn once' comment above rings true to to me, similar to Slackware in the Linux world.
I think that the appeal judge decided that
a) The insurance company bought stuff to use with the IBM software that didn't arrive
b) Some of that stuff could not be used with any other system
c) Therefore the cost of the subset of stuff that could not be repurposed could be added to the loss figure as 'wasted expenditure'
d) The clause in the original contract that IBM claimed disallowed 'wasted expenditure' from their liability for non-completion in fact didn't.
PS: Was it IBM that appealed against the original judgement?
"The best that could be hoped for was some proportionality to any such change in the law - i.e. the loss of protection applies only to individuals working in a giga-corp."
I thought that the article suggests that the DoJ are asking the court to decide? Along the lines that a quick cc to a lawyer with no specific question and no trace of any reply is not sufficient to claim the protection. More specific actual questions together with actual replies would still be protected therefore.
Icon: what do I know?
@YAAC
Some numbers: HS2 train frequency 14 trains per hour, 1000 seats per train so 14000 people per hour. Say arrival time in London is 9am. Suppose so early birds go for 7am start. 28000 people extra in a city of 1 million (plus another million in the hinterland) isn't going to make a *huge* amount of difference.
What have I got wrong?
PS: 45 min into Birmingham at 5pm then an hour or so to get anywhere else more than 5 miles outside the city. Local transport isn't great at present. Metro extension might help.
"Solve everyone demanding a 4 bed with a large garden and build liveable flats or medium density"
@ac2022-03-01_16:42
Come to Birmingham (within the Ring Road).
Loadsa new low rise flats + medium density housing. Google Port Loop (Urban Splash development) Park Central (Crest Nick) and St Lukes Road (Barrat, not as bad as it sounds). Plus stuff happening out by the Commonwealth Games site (Perry Barr) and Hipsta Urban speculative build stuff in the actual centre of the city. If you have serious money there are a handful or several new terraced houses in the Jewellery Quarter, plus various smaller scale developments.
What we actually need though is more council/housing association housing. Urgently.
Just while we are swapping obscure mirrors...
http://www.slackware.no/
These people do a 32bit build of the full *current* install iso each Tuesday. NB it isn't hybrid. You have to burn it to an optical disk.
(Both the main ones are going fine from here as well as slackware.uk and mirrorservice.)
I did not download Slackware over wet string dial up modem and install it from punched cards floppies back in the Edwardian era early 90s. I started using Slackware in 2014 with 14.1.
You get a large system (compilers, build tools, server stacks and databases and so forth, scripting, document preparation tools) and a full desktop with applications. Most applications are supplied as released by the upstream project, often with full documentation.
This base is maintained by PV with input from the wider team with updates made available. You get to decide if and when to install them. There is no automation (unless you add it), the system is configured mostly with scripts in /etc. I even understand a few of them.
No drama, works for me.
@Chz, @ibmalone, @Charlie, @Anonymous
Thanks for taking the time to reply.
@Chz: if any of the departments/groups have managed to get separate Google Workspace oganisations recognised in your University (and so multiple 100Tb allocations with different namespaces for users &c) I think many might find that useful information. PS Liking sending the students off to OneDrive
@ibmalone: glad to hear that the idea that data has a life beyond the project and is a deliverable in its own right is at least being thought about
@Charlie: Sharepoint has its moments speaking as an end user. I like the especially like the OFSTED proximity function (3 day notice of inspection received => authentication becomes unavailable for 12 hours or so, never fails :-)
@Anonymous: aha, yes I recognise that one from four decades ago (not winXP then of course...)
If anyone else has managed to get a Google Education/Workspace account recognised for an org that is not a school/University but is demonstrably a non-commercial org working in education I'd be interested in a pointer.
When I read the headline, I was thinking of schools in the pandemic &c here in the UK and thought OK 100Tb might do for homework upload and all at a stretch (2000 pupils in an 8 form entry secondary school = something like 10Gb/student + teacher accounts, class videos and all)
University researchers using the system to dump out petabytes of research data does strike me as being cheeky: perhaps research grant applications should include data lifecycle plans and salary for programmer(s) in these days of reproducible research?
I also scanned the eligibility requirements: geared to US. I'm guessing UK is basically for .ac.uk and .sch.uk domains. I'm wondering if there is a 'community organisation' route to eligibility akin to the US 'homeschool coops'?
Anyone got any direct UK experience?
Not sure but I think the idea is that the user could have been asked during the initial phone call to check the log.
Then the Windows disk full errors would become obvious when the user attempted to load the log or the last entry in the log would be something like 'save file failed disk full' or whatever.
"So you honestly think that Microsoft should stop people from being allowed to mess with the equipment that they bought and paid for?"
Sigh.
I understand and agree with this reaction to the extent that I personally have decided not to use Windows but to stick to (slackware) Linux with OpenBSD as a fallback position.
But.
Software in the age of ubiquitous always-on distributed network connections is porous - think colostomy bag - and the writers of commercial software will want to move to charging rent rather than selling stand alone applications. So the boundary between our rights as customers/tenants and the software provider's rights as landlord is fractal and shifts. In the same way the dividing line between what is hardware (which I own) and software (which I have a licence to use subject to terms and conditions - e.g. a tenancy agreement) is pretty vague. Is the bios/boot code hardware or software?
Also.
Ordinary people (the ones who never come here) have problems managing a user name and a password and have problems saving files, let alone getting their head around processes on a CPU &c. Perhaps file sync to data stores in the 'cloud' (i.e. landlord's servers) is better for ordinary people.
I dunno - views?
Icon: I accepted some decades ago that I would always be the one outside in the street looking in on the feast.
@idiot taxpayer
I recollect fax-spam was a problem back pre-millennium. Is that still a thing?
I actually miss the days when I could run everything off a phone and a fax (voluntary sector).
I'm assuming that your business must be in a good niche and that you have long term customers. Best of luck with it.
@Grunchy and all
Lloyds is an insurer of last resort in my understanding. If I have this right, when things go really pear shaped, they will cover the insurance companies that cover the risk. As such there is a need for resilience (the real kind not the kind management types talk about) in the face of unusual events.
AWS is successful but has a record of being a point of failure that affects many services. It has been around for what, 16 years or so? I grew up in a coastal place that has a major storm that causes significant damage to buildings once per saros roughly (as a result of tidal patterns interacting with weather patterns in autumn and the ocean swell). Hummm
OA has a paragraph that ends "...while automating manual processes.".
As always, one wonders if these are the process that the senior managers think are used, or the processes that are actually used on the ground?
Best of luck. One to watch because of the risk of systemic failure just at the point where you don't need it.
"Now, about that official Linux version..."
...which is very unlikely to be installed by me.
The last thing I want is some kind of conditional syncronisation of my whole ~/home/user where (if I have understood correctly) files may or may not actually reside in my home directory once edited.
Best of luck all those who help confused people trying to work out why they can't find a file.
@demon and all
Below is a quote from the OA
"If you're still using OpenOffice, don't. It's basically dead. Download LibreOffice, uninstall OpenOffice, then install LibreOffice instead. It's completely compatible because it's the same program, just a more modern version – smaller, faster, less buggy, and more secure."
I'd agree with the first sentence of the quote for new users or users without a large number of files produced using [Star |Open]Office.
Draw and Impress files with drawings produced in OpenOffice and opened in LibreOffice may warn about new settings and an incompatible change depending on the OpenOffice version that they were last opened with, so users like me who have files going back to OpenOffice 1 days need to step carefully. I'm still scoping this one out...
Icon: for all who contribute to either project. The more languages the merrier.
@Idiot and @Boris
The 'Limitations' section of the Nature article reporting the research makes interesting reading. If I have read this correctly, the study omitted social media.
The authors of the study also provide the original dataset, so some attempt at binning by age may be possible although the statistical errors in cross tabbing may be large as N=2k or so for the whole study.
In the education industry, the interaction between options baked into standard operating system images, users with limited rights unable to change option settings, and power management hints to external monitors (e.g. screen projectors) causes a never ending stream of multilayered fun.
Add in OS updates and separate application auto-updates for bonus laughs.
@Terry and all
Its a game of exponential growth (with the ultimate roll off in growth rate)
Even a small reduction in the concentration of virus around a person will have an effect on the likelihood of that person becoming infected and then infecting others. Compound that small advantage through a chain of 10, 20, 50 or 100 contacts and the difference in infection frequency at the other end will be dramatic. So even if linen masks are not amazingly effective, their use at population level can have a powerful effect. Hence as mentioned up the thread people in countries who experience regular outbreaks of respiratory diseases naturally wear masks in confined crowded places when there is an infection active. No drama, no politics, just common sense.
I'd love to know if anyone has done anything like an Onsager model for the interaction graphs of (say) 30 people getting off a commuter train - track them forwards for a week. Say one was infected and everyone stayed on the train for a half hour ride...
The models currently used appear to be based on averages treated using continuous differential equations (thermodynamics if you like). It would be nice to try a detailed simulation of individual interactions and build up to a thermodynamic limit just to get an idea what shape the resulting distributions had. Perhaps a use for all the processing power that is currently used to guess what things I bought last week and show me more adverts about identical things...
Back on thread: no issues with wifi from about Ubuntu 6.06 onwards. Intel cards/Thinkpads.
Just what I do...
/etc/hosts file set up with one of those curated hosts files that block many of the add serving domains
Chromium and Firefox kept for Websites with significant program-based content that I actually want (generative art sites and so forth), 'tube &c, along with private browsing windows for online banking and govt web sites.
Seamonkey for random stuff (bit slow and some 'modern' css does not work too well)
Something like uzbl or surf invoked for news/information sites. Can be invoked with javascript off.
Basically moving to a LoFi Web which is mostly text and pictures.
"Red Hat isn't "legally required" to release any code other than that covered by the GPL, and they're only required to release that code to their customers, not to everyone."
That was my understanding as well.
The Red Hat project has contributed mightily to the Linux kernel over the years and to many applications that provide the Linux operating system functionality.
I suspect that part of the pearl-clutching reaction to Red Hat rendering CentOS 8 EoL years earlier than the published 10 year lifetime is a realisation that Red Hat's new owners could turn off the oxygen tap at any point by simply not publishing the source code.
A bit like a cartoon character that runs off a cliff but keeps on running along in the sky until they look down...
A public company that makes arbitrary changes to its own published support roadmap part way through a cycle for no apparent reason in terms of profit or turnover must expect to find less trust being placed in its future roadmaps, and thus erosion of the value of its core OS.
As I have posted in other threads, if CentOS 8 had been allowed to reach EoL with a definite shift to Stream from 9 onwards there would have been time to plan for this and probably less consternation.
As it happens, Fermilab/CERN are looking at CentOS 8 Stream which will mean plenty of contributions or at least bug reports...
In the HEP community last time (RedHat Linux to RHEL, free to download to paid subscription, and suddenly announced) the total licencing cost for the cores in use at Fermilab/CERN alone was many (like 10^x) times the cost of hiring staff to recompile the SRPMs. Hence Scientific Linux.
Budgets are set on timelines not far off an RHEL version lifetime. A step function increase when not expected is difficult to deal with.
@fredbloggs
You missed out a category of user (or 'consumer' in RHELspeak) who do need stability &c but who don't have huge budgets, although they do have some budget for IT people.
Here is the most recent activity I could find...
https://www.mail-archive.com/scientific-linux-users@fnal.gov/msg06598.html
https://www.mail-archive.com/scientific-linux-users@fnal.gov/msg06731.html
So Red Hat will see some contributions to stream 8 from a user group who do use quite chunky hardware and who do have a number of installed nodes. There could well be a jump to Alma/Rocky or whatever after that.
@Reg and all
Don't forget PUIAS Linux, now Springdale Linux. Has been around a long time and is carrying on.
@Phones...
"...to a sudden panic for financial security when the government attitude switches tack to claw all the money back it is currently spending to keep people out of work"
What kind of timescale do you see for this bright and welcoming future? Might make the next (few) general elections interesting if timed incorrectly,
Icon: oldie, survived the Thatcher era mostly in one piece.
"However, many workers, including Brunel himself, soon fell ill from the poor conditions caused by filthy sewage-laden water seeping through from the river above. This sewage gave off methane gas which was ignited by the miners' oil lamps." --Wikipedia page on the tunnel
I stand corrected.
"The visiting US team unexpectedly announced that they were shutting down the whole UK operation, making everyone redundant."
@Blofeld's Cat
Shades of the Bournville factory in Birmingham shortly after a takeover of Cadbury's by an American food combine.
At least, after such a blow, people can hold their heads up high. No pretence of selection for redundancy on the grounds of poor performance or failure to adapt. Just a business decision about the presence of the owner's corporation in the territory.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
