* Posts by Paul Floyd

35 publicly visible posts • joined 22 Jan 2007

What can be done to protect open source devs from next xz backdoor drama?

Paul Floyd

It didn't make it into Fedora 40. Too many test failures.

Tired techie 'fixed' a server, blamed Microsoft, and got away with it

Paul Floyd

Restore from backup?

You've seen things people wouldn't believe – so tell us your programming horrors

Paul Floyd

Debian never fails to disappoint. The Purify condiitonal compilation should heve been a red flag.

The correct fix for the Valgrind errors was to

a) #include <valgrind.h>

and

b) change "#ifndef PURIFY" to "#if !defined(PURIFY) && !RUNNING_ON_VALGRIND

If adding a dependency on valgrind.h is too much to ask, Valgrind has a suppression mechanism for things like this.

Mozilla CEO quits, pushes pivot to data privacy champion... but what about Firefox?

Paul Floyd

Almost 7 million bucks compo for being CEO of an opensource foundation. Seems rather excessive to me, even by US corporate greed standards.

KDE 6 misses boat to make it into Kubuntu 24.04

Paul Floyd

And how many OpenBSD desktop users are there?

Windows 3.11 trundles on as job site pleads for 'driver updates' on German trains

Paul Floyd

Re: Improvement?

> replacement hardware -- Windows 3.1 doesn't run well on modern hardware

You are quite likely missing the point there.

I'd say that there is a good chance that this is running on old kit with something like a custom adapter card (perhaps as old as PC-AT format).

Windows 11 doesn't work well on ancient hardware.

BOFH: Monitor mount moans end in Beancounter beatdown

Paul Floyd

Has the PFY been replaced by the PYF?

Apple exec defends 8GB $1,599 MacBook Pro, claims it's like 16GB in a PC

Paul Floyd
FAIL

Nah

Generally I would expect compiled code on the amd64 architecture to be smaller than anything ARM based.

I don't think that Apple can claim with any honesty that current macs have the same or less bloatware than previous generations (my rough estimation of bloat based on the number of processes that are running after booting is that it has quadrupled over the last 15 years).

It could be that Windows and the Unix-likes are horribly ineffecient and that the super smart Apple engineers have managed to fit a quart in a pint pot. But I don't believe that either.

My theory is more prosaic. Apple can't be arsed with the cost and effort of modifying their designs and are happy enough overcharging for their current offer.

Phones and tablets are a diferent story but you can't extrapolate from a phone to a desktop.

Oracle pours fuel all over Red Hat source code drama

Paul Floyd

Re: Opensolaris anyone?

Why did Solaris take off in a big way? SunOS had been doing well enough on M68K, but when SPARC was released it wiped the floor with the competition. Still, relatively low volume and high profit. Fast forward about 10 years to the late 90s. Chips like the Pentium III have now caught up with the SPARC (chips like the Ultra SPARC II) but are cheaper.

Skip forward another 5 years to 2003 and AMD releases the Opteron. By now the SPARC is no longer competitive on cost/performance. But at that time Sun had a big opportunity. Sun already had several years of experience of 64bit OSes, the V9 SPARC architecture having been released about 8 years earlier. Sun also released a 64bit amd64 version of Solaris 10 in 2005.

Though Linux had already supported amd64 for a few years my memory was that it was all very flakey. Sun did push fairly hard for amd64, but also they had done some serious damage as already noted when they dropped Solaris 9 x86 for a short while. Even though Sun was selling Opteron workstations and servers, most customers were using them for Windows or Linux. And though the writing was on the wall, Sun was still making most of its money from SPARC and couldn't bring themselves to switch focus to amd64. OpenSolaris was too little, too late. And finally along came My Little Pony and finished the job off https://www.theguardian.com/technology/blog/2010/feb/04/jonathan-schwartz-sun-microsystems-tweet-ceo-resignation.

Quirky QWERTY killed a password in Paris

Paul Floyd
Facepalm

Recipe for RSI

French typists may be faster

But for software development (especially the C family of languages) the keyboards are a nightmare.

# - right next to the enter and shift keys on a UK board, AltGr-3 on a French board

\ - next to the left-shift on a UK board, AltGr-8 on a French board

Square brackets and braces aren't on adjacent keys, instead they are symmetric bu spread over the number row.

Inclusive Naming Initiative limps towards release of dangerous digital dictionary

Paul Floyd

It can be worse in other languages.

English is mostly un-genendered when it comes to (pro)nouns and articles. Inclusive French seems to be getting more common. Stuff like 'iel' instead of 'il' or 'elle', and making a mish mash of gendered nouns. That's not so bad when it's just an 'e' at the end (so etudiant-e) but it looks a lot worse when the masculine and feminie versions differ more. For instance "Conducteur·rice de train" (train driver, copied and pasted from this ad https://metiers.siep.be/metier/conducteur-conductrice-train)

Paul Floyd

Re: Nope.

There's already been a bit of controversy when the image for Beastie changed from a cartoon demon wearing daps to being a much more abstract red sphere with pointy ears.

Who writes Linux and open source software?

Paul Floyd

Not all big corps contribute

My experience is that Apple is conspicuous by their absence in opensource projects. Sure, they contribut to LLVM, but I feel they prefer to take than to give.

Also IMO ARM doesn't pull their own weight.

Microsoft begs you not to ditch Edge on Google's own Chrome download page

Paul Floyd

Trust them as far as I can throw them

Do any of these big corps really believe that anyone trusts them?

They are hardwired to spin their corporate bollox.

The sad state of Linux desktop diversity: 21 environments, just 2 designs

Paul Floyd

Oberon System

Back in the day there were various ports of the Oberon System. The original GUI (S3) is a tiling WM with a rich set of mouse 'interclicks' (a later non-tiling WM was also developed, V4).

Eclipse boss claims Visual Studio Code is an open-source poseur – though he would say that, wouldn't he?

Paul Floyd

Re: Netbeans ftw

Speaking as an ex-netbeans user. When is cnd going to be updated (or replaced) on the current NB platform, preferably with the most recent C++ support?

Post-netbeans I mainly use Qt Creator. Works quite well, though the git integration is rather cheesy.

Oracle creates new form of free Solaris

Paul Floyd

Re: OpenIndiana

OpenIndiana - not really. Due to lack of available hardware (and maybe people able/willing to do the work) OpenIndiana Hipster no longer supports SPARC.

https://www.openindiana.org/es/documentation/faq/#does-openindiana-provide-a-sparc-release

There seem to be a few (more) obscure spins that still support SPARC.

Intel energizes decades-old real-time Linux kernel project

Paul Floyd

I think that you are confusing "fast" with "guaranteed response time".

These couldn't wait for Patch Tuesday: Adobe issues bonus fixes for 92 security holes in 14 products

Paul Floyd

Re: RE: Give it time

And what language might that be?

Spot the dog? No, we couldn't either because Spot is a robot employed by United Kingdom Atomic Energy Authority

Paul Floyd

Hopefully it will last longer than the robots used at Chernobyl. I suppose that fido won't be exposed to such extreme levels of radiation.

Waterloo! Windows defeated, your sign is screwed. Waterloo! Promise to bork you forevermore

Paul Floyd

This story (the Waterloo insult one) seems to have spread a fair bit on the 'net. Another example of crowd-dumbing. It seems as though no-one knows how to spell the poor chap's name.

Monster magnet in my pocket: Boffins' gizmo packs 45.5-tesla punch and weighs just 390g

Paul Floyd
FAIL

Re: The numbers are not tautological.

It's a coil. An inductor. One of the 3 types of passive circuit element that exhibit electrical impedance.

Quicky refresher, for ideal components and steady state. A resistance lets current flow through it, proportional to the voltage. A capacitor will have current flow until it is charged and then it will have a voltage across it but 0 current. In inductor will have a voltage across it until the magnetic field builds up when it will have current but 0 voltage across it.

These numbers for coil voltage and resistance are pretty close to zero are pretty close, which is why there is such a high current and magnetic field.

Paul Floyd
Headmaster

Re: When will I be able to put it on my fridge door ?

You don't want to put it on your door. You want it to run your fridge.

https://en.wikipedia.org/wiki/Magnetic_refrigeration

As Tesla hits speed bump after speed bump, Elon Musk loses his mind in anti-media rant

Paul Floyd

Re: unexpected honesty

What about bankers?

Vibrating walls shafted servers at a time the SUN couldn't shine

Paul Floyd

Re: I call bull

For the electronics, it's not just a question of the field strength. It's more a question of the rate of change of the field.

For the hard disk, the write head is **very** close to the magnetic material. The magnetic filed falls off with the cube of the distance. So I guess it would have to be a fairly enormous MRI-style superconductor magnet that would be required.

Paul Floyd

Re: I call bull

I suggest that you take a refresher on electromagnetic induction.

Labour says it will vote against DUP's proposed TV Licence reforms

Paul Floyd

Re: News to me

Re: German TV license

It's much the same in France where we have the redevance tele which gets collected with your council tax. You can opt out if you don't have a goggle box. You don't have to pay of use the internet though.

Dark times for OmniOS – an Oracle-free open-source Solaris project

Paul Floyd

Re: It was designed to fail

You are seriously misinformed. If you wind your clock back to 2005 then the picture is very much different. Back then, Linux was pretty much only 32bit, and the 64bit version was not yet fit for purpose. Sun, by that time had a lot of experience of 64bit Solaris on SPARCV9. So when Solaris 10 came out, the 64bit amd64 version was probably the best OS for the platform. Remember, Intel was still sailing towards the iceberg on the Itanic, and AMD was trying to steal a march with Opteron. AMD stumbled with Barcelona and Bulldozer, Intel struck back with em64t, Sun failed to get much traction for Solaris amd64 even on its own hardware, OpenSolaris was too little, too late and the rest is history

Emulating x86: Microsoft builds granny flat into Windows 10

Paul Floyd

Re: Cart before the horse

Apple did it twice even. Moto 68k (the original Macs) to PowerPC with 'fat binaries' and then PowerPC to Intel with 'universal binaries'. This has the added builtin marketing feature that when the old platform is finally deprecated they can claim to free up disk space.

Anyone using M-DISC to archive snaps?

Paul Floyd
Boffin

Data for the fourth millenium

Whilst I've not worked in the field for a long time, my PhD was on pit-forming mechanisms in dye-polymer optical storage.

I'm not too impressed by the FAQ. "the M-DISC™’s data layer is composed of rock-like materials known to last for centuries". The Wikipedia M-DISC article is a bit better.

Normally optical media uses a polycarbonate substrate. This has pretty good optical and physical properties (and is dirt cheap). However it is somewhat hygroscopic, and when under stress, the optical properties are less nice. On top of this, CD/DVD/BD-LTH have an organic dye-polymer layer (100nm or so thick), a layer of aluminium (50nm) and a layer of laquer (10um) on the label side. 10 microns of laquer doesn't offer much physical protection. Back on the polycarbonate substrate side, small scratches tend not to be too much of a problem because the disk is fairly thick (1.2mm) and the incident laser has a fairly high numerical aperture (i.e., it's still a fairly wide spot when incident on the disk, but converges at a high angle to the dye/reflective layers [as high an NA that you can get with a lens that probably only costs 20p]). Also the Reed-Soloman ECC does a pretty good job. I don't know if this has improved with DVD or BD.

I doubt that the M-DISC deviates that much from the CD/DVD structure. The main difference seems to be in the recording layer, which is more like BD. I'm not sure about the thousand year claim, but if you take care not to scratch the disks, write on the label with a felt pen that won't dissolve the laquer and keep them somewhere fairly dry, then I reckon they'll last well compared to alternative data storage media.

Silicon daddy: Moore's Law about to be repealed, but don't blame physics

Paul Floyd

Many issues

There are many issues involved in continued die shrinkage. Just to list a few. There's the problem of making masks. Currently there are large sets of design rules in order to be able to create masks with dimensions using light that is of a much larger wavelength. People have talked about moving to shorter wavelengths, but again there is a big economic barrier. Next there is the issue of what exactly scales. Back in the old days, you had 5V and you could just shrink the dimensions and nothing else. But then the electric field (voltage/distance) started getting too high, so the voltage had to start dropping. Second but, it couldn't drop as fast as dimensions shrank. This has a speed/power tradeoff, but basically silicon transistors don't work below about 0.6 to 0.7 volts (the threshold voltage where a transistor switches between off and on). High-k dielectrics were introduced to help with the electric field breakdown issues. Next there is the problem of variability. One of the important aspects of IC design is that while it isn't easy to exactly control the parameters of the transistors (e.g., to have precise resistances and gains), it used to be the case that transistors physically close on the die would be very closely matched in characteristics. When you scale down to small numbers of atoms, then each transistor has much more statistical variation. This makes design much harder.

Paul Floyd

This of course assumes that shrinkage leads to continued reduction in power consumption.

Boffins, Tunnel Tigers and Scotland's world-first power mountain

Paul Floyd
Pint

See also: Dinorwig

I guess that Cruachan is very similar to Dinorwig, which I visited as a student when I was studying Electronics in Manchester. A pint in memory to the ones that we downed in Chester on the way back.

eBayer mails UK lad £44k

Paul Floyd

How much space?

In the original article, it's 65KEuros. If they're 500Euro notes, that'd be 130 of them. The size of a small paberback. 200 and 100Euro notes would probably also be feasible.

Is Solaris really a bright choice for developers?

Paul Floyd

Some Corrections

I don't think Mike Wright has much experience with Solaris, to be honest. For free you can get the OS and critical and securty patches, and any such patches that are dependencies. You have to pay for the rest. With Red Hat Enterprise Linux or SuSE Enterprise Linux you pay for the OS and annual support. If you want compare it to an all-free Linux like Fedora Core, then you should compare it to Solaris Express. That's free and has updates roughly every month (though no patches of any kind).

sh as the only shell? Of course not. Personally I use ksh, but all of the shells are there. sh may be limited, but it's not as buggy as bash. Since Sun puts a lot of emphasis on compatibilty, there is no way that they will ever replace the Bourne shell with a symlink to bash.

From what I've seen, it's not just on high end hardware that Solaris gives Linux a run for its, err, money.